Pharmacologic Augmentation of Computerized Auditory Training in Chronic Psychosis: Preliminary Findings From a Single-Site, Double-Blind Study.

Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Schizophr Bull . 2025 Apr 14;51(6):1529–1544. doi: 10.1093/schbul/sbaf015 Search in PMC Search in PubMed View in NLM Catalog Add to search Pharmacologic Augmentation of Computerized Auditory Training in Chronic Psychosis: Preliminary Findings From a Single-Site, Double-Blind Study Neal R Swerdlow Neal R Swerdlow 1 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 2 VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States Find articles by Neal R Swerdlow 1, 2 , Joyce Sprock Joyce Sprock 3 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 4 VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States Find articles by Joyce Sprock 3, 4 , Francesca Li Francesca Li 5 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States Find articles by Francesca Li 5 , Jenny Min Din Jenny Min Din 6 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States Find articles by Jenny Min Din 6 , Jessica Minhas Jessica Minhas 7 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States Find articles by Jessica Minhas 7 , Jo Talledo Jo Talledo 8 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States Find articles by Jo Talledo 8 , Yash B Joshi Yash B Joshi 9 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 10 VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States Find articles by Yash B Joshi 9, 10 , Juan L Molina Juan L Molina 11 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 12 VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States Find articles by Juan L Molina 11, 12 , Bethany Nordberg Bethany Nordberg 13 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States Find articles by Bethany Nordberg 13 , Kevin Ing Kevin Ing 14 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 15 VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States Find articles by Kevin Ing 14, 15 , Michael L Thomas Michael L Thomas 16 Department of Psychology, Colorado State University, Fort Collins, CO 80523, United States Find articles by Michael L Thomas 16 , Gregory A Light Gregory A Light 17 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 18 VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States Find articles by Gregory A Light 17, 18, ✉ Author information Article notes Copyright and License information 1 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 2 VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States 3 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 4 VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States 5 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 6 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 7 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 8 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 9 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 10 VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States 11 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 12 VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States 13 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 14 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 15 VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States 16 Department of Psychology, Colorado State University, Fort Collins, CO 80523, United States 17 Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States 18 VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States ✉ To whom correspondence should be addressed: Gregory A. Light, Department of Psychiatry, UCSD School of Medicine, 9500 Gilman Drive, La Jolla, CA 92093-0804, United States ( [email protected] ) Collection date 2025 Nov. © The Author(s) 2025. Published by Oxford University Press on behalf of the Maryland Psychiatric Research Center. All rights reserved. For commercial re-use, please contact [email protected] for reprints and translation rights for reprints. All other permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site—for further information please contact [email protected]. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model ( https://academic.oup.com/pages/standard-publication-reuse-rights ) PMC Copyright notice PMCID: PMC12597525  PMID: 40227151 Abstract Background Computerized auditory training (AT) modestly improves symptoms, cognition, and functioning in schizophrenia. We assessed whether d-amphetamine (AMPH) or memantine (MEM) can enhance gains from 30-h of AT. Methods Antipsychotic-medicated individuals with chronic psychosis ( n = 68; mean age 47.03y; M:F = 39:29) completed up to 30 AT sessions (2-3/week; n = 50 completed 30 sessions) in 3 groups: “AMPH group” (AMPH (5 mg po) 1-h before each AT session); “MEM group” (titrated to 10 mg MEM bid and maintained that dose throughout training); “PBO group” (PBO dosed identically to either AMPH or MEM). Primary (PANSS total, MCCB Composite, WHODAS) and secondary (PANSS positive, PANSS negative, YMRS, PHQ-9, PSYRATS) outcome measures were acquired at baseline, after 10, 20, and 30 AT sessions, and 12 weeks post-training. Pill identity (active/PBO) was blind to subjects and staff. Results Marginally significant between-group gains for AMPH vs PBO were detected for one of three primary outcomes (WHODAS, P =.050; but not PANSS total or MCCB Composite), and for 3 of 5 secondary clinical outcomes (PANSS positive, YMRS, PSYRATS, P ’s≤.027–.049). Within-subject gains over time were detected for primary and secondary clinical measures for AMPH ( P ’s≤.014–.004) and MEM ( P ’s≤.02–.001) groups; some of these would not survive conservative correction for multiple comparisons. No measures detected symptom worsening; treatment satisfaction exceeded subjects’ expectations. Conclusions Results are mixed; drug-associated gains in several measures vs PBO suggest that these regimens may augment AT-induced functional and clinical improvement in psychosis patients, independent of changes in neurocognition. Assessment in larger samples seems warranted. Keywords: Amphetamine, auditory training, memantine, pharmacologic augmentation, schizophrenia Introduction In schizophrenia (SZ) and related disorders, neurocognitive impairment is a determinant of psychosocial disability. 1–3 While antipsychotics (APs) marginally improve neurocognition, 4 specific cognitive therapies enhance neurocognition and outcome in SZ patients, with effect sizes of d ≈ 0.40 vs treatment as usual. 5–9 Studies document the safety and efficacy of cognitive therapies in SZ, with benefits sometimes lasting years. 6 , 10–13 In computerized Auditory Training (AT) 14 patients perform progressively more difficult auditory-based learning tasks to improve processing in auditory, attention-related, and working memory systems. After 30-50 hours (h) of AT, some SZ patients have large effect-size gains in auditory-dependent cognitive domains and global cognition that persist for at least 6 months. 14–20 Some multi-site findings show significant gains in MATRICS Consensus Cognitive Battery (MCCB) composite and verbal learning scores after 20 h of training. 21 Nonetheless, many SZ patients fail to benefit from AT ( d ≤ 0.2) 22 , 23 or show limited clinical gains even after 100 h of training. 15 Given the high rate of AT “non-response,” attrition rates that often exceed 30% 20 , 24 , 25 and modest effect sizes, the costs, and logistical impediments associated with getting impaired SZ patients to complete a course of AT can be prohibitive. Clinical gains are not always reported with AT. 14 , 15 , 23 Some studies identify functional gains in SZ patients after AT (eg, 26 ); in other cases, individual clinical scales detect at least small-to-medium effect size gains. Thomas et al. 17 , 18 reported that in treatment-refractory psychosis inpatients, 30 h of AT yielded small-to-medium effect size reductions in positive symptoms ( d = 0.46), and gains in psychosocial treatment engagement. Loewy et al. 20 reported reductions in total ( d = 0.31) and positive symptoms ( d = 0.54) after 40 h of AT administered remotely. Scoriels et al. 24 reported larger reductions in positive symptoms ( d = 0.82) with 40 h of AT, and more generalized reductions in psychopathology scores with 40 h of computerized visual training. Still, the literature suggests that AT-induced gains in neurocognition are not always accompanied by symptomatic improvement. This is not unique to AT: a meta-analysis of “drill-and-practice” computerized cognitive training for psychosis—including AT and other training programs involving repetitive practice—identified clinical gains limited to small effect sizes in positive ( d = 0.31) and depressive symptoms ( d = 0.37). 27 There may be ways to amplify the benefits—cognitive, clinical, and functional—of AT. We proposed a strategy of Pharmacologically Augmented Cognitive Therapy (“PACT”) for SZ, 28 , 29 in which pro-cognitive drugs specifically augment the gains from cognitive therapies. These drugs would not replace APs, which remain essential for limiting psychotic symptoms that impede a patient’s ability to participate in cognitive rehabilitation. PACT requires the availability of drugs that enhance AT effects in SZ patients, yet trials of pro-cognitive agents per se in SZ have largely yielded negative results. 30–35 “Negative” studies typically were not conducted in the context of cognitive therapy, and drugs designed to enhance specific domains of cognition, e.g., attention, might not yield clinical benefits unless paired with cognitive interventions that place demands on those domains. In many “negative” studies, candidate pro-cognitive agents were simply added to a passive daily medication regimen, without any new cognitive “load” (i.e., increases in cognitive exercises/engagement). Proof-of-concept for the PACT approach of pairing drugs that have a specific pro-cognitive mechanism with a therapy that demands that mechanism is found in the use of pro-extinction drugs to enhance the impact of exposure therapy for anxiety disorders. 36 , 37 We assessed the ability of 2 medications to enhance neurocognitive, clinical, and functional measures in psychosis patients undergoing 30 h of AT. The psychostimulant, d-amphetamine (AMPH), and the uncompetitive NMDA receptor modulator, memantine (MEM) acutely enhance aspects of auditory processing as well as auditory learning in psychosis patients, assessed with a frequency modulation task that is part of a well-characterized AT suite. 38–40 Because the neurocognitive, clinical, and functional gains from AT presumably reflect the learning that occurs during AT, we hypothesized that, administered throughout training, these agents would also enhance the benefits of a full course of AT. Methods Recruitment and Screening Two separate protocols were approved by the UCSD IRB (protocols #191811, #201502) and posted on ClinicalTrials.gov. AP-medicated (stable regimen > 30 d) patients with a primary diagnosis of SZ or schizoaffective disorder (SAD) were phone- or field-screened for inclusion/exclusion criteria ( Table S1 ). Appropriate subjects ( n = 120; Figure 1 ) came to UCSD for consenting and a comprehensive screening and diagnostic assessment ( Table S2 ; including a modified version of the Mini-International Neuropsychiatric Interview (MINI 41 )) (“T 0 ”). The original study design included separate, parallel assessments of AMPH (vs placebo) and MEM (vs placebo) effects on AT-associated clinical, neurocognitive, and functional gains. However, these designs pre-dated the pandemic, and between the date that they were proposed (10/2019) and the project start date (8/2020), our institution implemented a pandemic-related shut-down of clinical research activities. Even with the gradual resumption of research activities, recruitment lagged significantly behind target levels. As a result, we made 3 pragmatic adjustments to the original study designs: (1) the 2 parallel studies were combined and placebo groups were pooled, to yield a 3-group design (AT + AMPH, AT + MEM, and AT + PBO); (2) subject randomization changed from 1:1 to a 2:1 active: placebo ratio; and (3) adjustments were made to the planned statistical analyses, appropriate to the reduced sample size. Figure 1. Open in a new tab CONSORT flow chart shows disposition of 120 consented subjects, 68 randomized subjects, and 50 study “completers.” Abbreviations: W = withdrew from study; E = excluded from study; tox = urine toxicology screen positive for methamphetamine (excluded after 3 rd consecutive weekly positive result; see text); SA = suicide gesture/attempt (subject drank tea brewed from flowers, was seen at an emergency room and was discharged without further intervention; see text). Comparisons of baseline clinical scale scores in PBO group subjects allocated to the AMPH study protocol vs the MEM study protocol (i.e., who received PBO dosed identically to either AMPH ( n = 10) or MEM ( n = 17)) revealed no significant group differences on any measure. Based on this, for all analyses, PBO groups from the 2 study protocols were pooled to form one unified PBO group ( Figure 1 , bottom). “Arm” assignment (AMPH vs MEM) alternated unless the subject met study-specific exclusion criteria ( Table S1 ). After initial screening (“T 0 ”), subjects returned twice (“T 1 ” and “T 2 ,” ≈ 7d apart) for biomarker assessment after challenge with PBO (Test 1) or test drug (PBO, AMPH 5 mg po or MEM 20 mg po: Test 2) ( Supplemental Methods and Table S2 ), and then entered the “treatment phase,” completing up to 30 AT sessions (≈ 2-3 sessions/week) in 3 groups: “AMPH Group” received AMPH (5 mg po) 1h before each AT session; “MEM Group” titrated over 4 weeks to 10 mg MEM twice daily before starting AT and maintained that dose throughout training; “PBO Group” received PBO dosed identically to either AMPH or MEM. Pill identity (active vs PBO) was blind to subjects and staff. AT consists of 7 computerized exercises delivered on standardized laptops and headphones. Collectively, these exercises target learning mechanisms involving auditory perception and processing speed (Sound Sweeps, Fine Tuning) and auditory memory (Syllable Stacks, Memory Grid, To-Do List Training, Rhythm Recall, Hear-Hear). Training is structured into blocks that deliver stimulus sets with varying temporal and psychophysical parameters to allow continuous learning and improvement. Blocks consist of 10-35 adaptive trials where the subject’s progression depends on their performance. Exercises apply an n-up/m-down algorithm to responses to estimate psychophysical thresholds while ensuring that participants remain engaged and challenged at an appropriate level (~80% accuracy) as their abilities improve. AT details are found in. 14 , 15 , 17 , 18 , 42–50 On average, subjects completed 2.44 AT sessions per week and spent 11.48 weeks in treatment. Schedule of Measures The effects of AT and medications on neurocognitive performance were assessed using the MCCB Composite T-Score, which is consistently reported to be sensitive to AT 14–26 ; performance was assessed at T 0 , and 1-2 days after 10, 20, and 30 training sessions (“P10,” “P20” and “P30”) and 12 weeks post-training. Clinical and functional outcome measures were acquired at T 1 (“baseline”), P10, P20, and P30 and 12 weeks post-training. Urine toxicology screens and Columbia Suicide Severity Rating Scales (C-SSRS 51 ) were performed at least weekly, prior to an AT session. A treatment satisfaction scale (100 mm line) rated expectations (at T 0 ) and actual experience of treatment (at P30) in 3 areas: “satisfaction,” “hard work,” and “worthwhile.” After P30, MEM group subjects (and corresponding PBO subjects) tapered-off their pills over 3 weeks; subjects from all groups returned to UCSD 12 weeks after P30, when outcome measures were reassessed to test the “durability” of benefits ( Table S2 ). Medication Protocols AMPH (5 mg po) and its PBO pills were administered by staff on-site 1 h prior to each AT session. MEM (10 mg bid) and its PBO pills were taken off-site; self-reported adherence to prescribed dosing of MEM (and its matched PBO) was confirmed at each visit by laboratory staff and/or subject self-report. Appropriate MEM/PBO dosing and prescription refills were also confirmed with staff at assisted living centers, who typically distributed MEM/PBO pills together with routine medications. The doses of AMPH and MEM were previously reported to enhance sensory information processing in SZ patients, and/or to enhance AT learning. 38–40 , 52 , 53 Statistical Analyses Based on known benefits of AT in psychosis patients, 14–26 no “placebo AT” condition (e.g., video game) was included, for ethical and pragmatic reasons. Since all subjects received AT, clinical gains over time in all groups might reflect, to some degree, the benefits of AT. Pharmacologic augmentation by AMPH or MEM would only be evident if/when benefits in “AT+active drug” groups exceeded those in “AT+PBO” groups. The primary goal of this study was to assess the clinical, neurocognitive, and functional gains associated with the addition of AMPH or MEM to a full course of AT. In total, there were 3 primary outcome measures and 5 secondary outcome measures. The primary clinical outcome measure was the Positive and Negative Symptom Scale (PANSS 54 ) total score; positive and negative subscale scores were secondary outcomes. Because they were highly correlated with total scores ( r = 0.92 at T 1 ), PANSS general scores are reported but are not outcomes. The primary neurocognitive outcome was the MCCB Composite Score. The primary functional outcome was the World Health Organization Disability Assessment Schedule (WHODAS 2.0, 12-item 55 ). Secondary clinical outcome measures also included the Patient Health Questionnaire (PHQ-9 56 ), the Young Mania Rating Scale (YMRS 57 ), and the Psychotic Symptoms Rating Scale (PSYRATS 58 ) hallucinations subscale. Clinical, neurocognitive, and functional outcome scores were treated as continuous variables and analyzed in 3 ways. First, they were submitted to linear mixed-effects (LME) models 59 for both between-group (interaction model: AT + AMPH vs AT + PBO; AT + MEM vs AT + PBO) and within-group (time model: baseline vs P30, all groups) analyses. LME analyses included data from all randomized subjects (including those who did not complete 30 AT sessions) and modeled random intercepts to account for within-subject dependencies. Second, because sample sizes fell well below target levels (see Methods and Discussion), to test the hypothesis that active drugs would augment the impact of a 30-session course of AT, the effect size (Cohen’s d ) of the change from baseline to P30 (“difference score”) was calculated for both between-group (AT + AMPH vs AT + PBO and AT + MEM vs AT + PBO) and within-group (baseline vs P30, all groups) comparisons. Between-group effect sizes were calculated as: d = [ ( d r u g s c o r e T 1 - d r u g s c o r e P 30 ) - ( P B O s c o r e T 1 - P B O s c o r e P 30 ) ] / S D w a v where SD wav is the weighted average standard deviation. Within-group effect sizes were calculated as: d = ( s c o r e a t T 1 − − s c o r e a t P 30 ) / S D a v ) where SD av is the average of the standard deviations of the T 1 and P30 measures. 60 Third, to disentangle the salutary contributions of AT alone vs AT + active drug to within-subject effect sizes, a “threshold” for positive active drug effects was set at “ d = 0.4 above PBO values” (see “Limitations,” #3): d (active drug) > d (PBO) + 0.4 ) ( Figure 2A ). Figure 2. Open in a new tab “Ideal” PACT model is represented. A. Baseline values of outcome measure (T 1 , open bar), expected reduction (improvement) in outcome measure resulting from 30 sessions (P30) of AT + PBO (lined bar), and the extra benefits of adding an effective “pro-learning” drug to 30 sessions of AT (solid bar). For the present study, in addition to linear mixed effect (LME) models and between-group effect sizes, drug-associated gains in clinical and functional measures were evaluated based on a “threshold” for added within-group benefits—above those observed with AT + PBO—that was set at d = 0.4. B. Change in outcome measure over time (from T 1 to P30) with AT + PBO (dashed line) and AT + active drug (solid line), showing the “threshold” of exceeding the benefits of AT + PBO by d = 0.4 at P30. Because clinical gains from AT + PBO were expected to be approximately d = 0.4, 17 , 18 , 20 this added threshold of “ d = 0.4 above PBO” for drug-enhanced AT meant that an estimated combined large effect size (at least d = 0.8) would be required for a drug effect to be viewed as meaningful. Outcomes using thresholds from d = 0.2 to d = 0.6 are seen in Table S6 . The use of effect size comparisons in addition to P -values is common in clinical trials research as a way to measure effects independent of sample size cf. 61–64 It is particularly useful when, as in the current study, sample sizes fall below targets identified based on traditional power analyses. Effect size “thresholds” have been used by others to argue that AT neurocognitive benefits exceed levels produced via PBO or practice effects (generally d ≈ 0.18). 65 Indirect indicators of clinical impact included subject attrition, C-SSRS responses, treatment satisfaction, and adverse events (AEs). AMPH group subjects were evaluated for abnormal involuntary movements (AIMS 66 ) and for deleterious effects of AMPH cessation (AMPH Cessation Symptom Assessment 67 ). Analyses included non-parametric (Chi-Square) and parametric (rmANOVA) tests. Linear mixed-effects models for outcome measures included all subjects who were randomized ( n = 68); df were estimated using the Satterthwaite approximation. Between- and within-group effect sizes for clinical, neurocognitive, and functional measures were calculated based on all subjects who completed 30 AT visits and P30 testing (“completers”; n = 50). For all analyses, α was set at 0.05; for analyses of the 5 secondary outcome measures, findings were also evaluated with a more restrictive alpha level (0.01) to correct for multiple comparisons. Supplemental Results describe more inclusive analyses (e.g., “last observation carried forward” (LOCF)), and analyses testing demographic and medication correlates of outcome measures, using simple regressions. Results Study Sample Table 1 shows characteristics of the randomized study sample ( n = 68). Subjects were typically high school-educated, never married, unemployed, residing in assisted living facilities, and had experienced psychosis for 2-3 decades. On average, they were taking 1.60 different APs, 1.68 other psychotropic medications, and experienced mild symptoms (average PANSS total score = 57.3). The 3 “drug groups” (PBO, AMPH, MEM) did not differ significantly in age, sex distribution, race, marital status, employment, smoking status, duration of illness, anticholinergic burden, or chlorpromazine equivalents. MEM subjects were slightly more educated than PBO subjects but not AMPH subjects. MEM subjects were more likely to carry a diagnosis of SAD Bipolar Type (SADBT) than were either AMPH or PBO group subjects, as individuals carrying this diagnosis were excluded from participation in the AMPH “arm” of this study ( Table S1 ); appropriately, MEM subjects were more likely to be prescribed a mood stabilizer. Baseline clinical scale scores and MCCB Composite Scores were generally comparable across the 3 drug groups. Baseline PSYRATS hallucination scores tended to be higher in randomized AMPH vs MEM and PBO group subjects ( P < .1). Table 1. Characteristics of Subjects Randomized to Drug Group ( n = 68) PBO group ( n = 27) AMPH group ( n = 18) MEM group ( n = 23) Age (mean y, range) 45.8 (22-65) 46.7 (34-59) 48.7 (21-65) Sex (M:F) 13:14 10:8 15:8 Smokers (%) 59.3 55.6 47.8 Diagnosis (MINI: SZ/SADDT/SADBT) 21/3/3 16/2/0 10/2/11 Ethnicity (Hispanic / Not Hispanic / Declined) 8/18/1 5/13/0 2/21/0 Race (Cauc/AA/Native American/Asian/Other) 15/4/1/1/6 9/5/2/0/2 15/6/2/0/2 Marital status (Mar/Never Mar/Div/ Wid) 1 3/20/4/0 1/12/4/1 1/17/4/1 Full employment (%) 14.8% 11.1% 13.0% Education (mean y, range) 2 12.1 (9-19) 12.8 (9-15) 13.4 (11-17) Duration Ill (mean y, range) 20.8 (2-39) 25.6 (1-35) 26.9 (1-47) Baseline MCCB Composite T-Score (mean (SEM)) 29.9 (3.2) 29.8 (5.0) 31.0 (3.4) Baseline PANSS Total Score (mean (SEM)) 58.4 (2.7) 57.6 (3.3) 56.7 (3.0) Baseline PANSS Positive Score (mean (SEM)) 14.0 (0.9) 15.1 (1.07) 14.5 (1.12) Baseline PANSS Negative Score (mean (SEM)) 15.1 (1.0) 14.4 (1.3) 13.7 (1.0) Baseline PANSS General Score (mean (SEM)) 29.3 (1.2) 28.1 (1.6) 28.5 (1.6) Baseline PHQ-9 (mean (SEM)) 6.3 (1.0) 6.9 (1.4) 8.5 (1.4) Baseline YMRS (mean (SEM)) 4.0 (0.8) 5.7 (1.2) 3.7 (1.0) Baseline BPRS subscale (mean (SEM)) 6.9 (0.5) 7.3 (0.9) 7.1 (0.7) Baseline WHODAS (mean (SEM)) 20.4 (1.5) 23.6 (2.3) 23.3 (1.7) Baseline PSYRATS hallucinations (mean (SEM)) 3 8.3 (2.1) 14.9 (2.8) 8.0 (2.3) CPZ equivalents (mg/d, mean (SEM)) 4 779.2 (165.2) 633.9 (76.1) 587.7 (108.4) Anticholinergic Burden Score (mean (SEM)) 5 6.6 (0.8) 5.8 (0.7) 4.9 (0.5) “N” taking regularly prescribed medications: First generation antipsychotics 6 4 2 Second or third generation antipsychotics 26 17 21 Clozapine 3 1 1 Antidepressants 15 8 10 Mood stabilizers 6 6 14 Anxiolytics 4 3 8 Anti-Parkinsonian drugs 7 4 3 Levothyroxine 2 0 2 Open in a new tab 1. Married/never married/divorced/ widowed. 2. F = 3.32, df 2,64, P < 0.045; MEM > PBO, P < 0.13 (ns). 3. F = 2.40, df 2,65, P < 0.1 (ns). 4. Self-reported doses, based on n = 25, 17, and 21 for PBO, AMPH, and MEM groups, respectively. 5. Based on summed total of 0–3-point scale for each medication (see Supplemental Methods ); F = 1.75, df 2,65, ns. AEs ( Table S4 ) were mostly mild, though one led to study exclusion: a MEM group subject drank tea brewed from flowers in a suicide gesture/attempt, was seen at an emergency room and discharged without incident. An AE was reported by 35.7% of PBO group subjects, 55.5% of the AMPH group subjects, and 21.7% of the MEM group subjects (PBO vs AMPH: χ 2 =1.76, ns; PBO vs MEM: χ 2 =1.19, ns; AMPH > MEM: χ 2 =4.98, P < .03). AEs among AMPH group subjects largely reflected: (1) 3 individuals each reported 1 night of restless sleep or insomnia, and (2) a greater occurrence of auditory hallucinations in this group consistent with their elevated T 1 (pre-AMPH) PSYRATS scores ( Table 2 ). Of the 876 C-SSRS among “completers,” there were 39 positive responses, reflecting positive response rates of 0.44%, 2.58%, and 8.82% for AMPH, MEM and PBO groups, respectively. Table 2. Outcome Measures: T 1 Scores (mean (SEM)) in “Completers” Measure PBO ( n =20) AMPH ( n = 13) MEM ( n = 17) PANSS total 59.80 (3.33) 55.62 (3.22) 54.24 (3.19) PANSS positive 14.95 (1.02) 15.23 (1.26) 13.41 (1.12) PANSS negative 14.65 (1.24) 12.85 (1.09) 13.65 (1.27) YMRS 4.65 (0.97) 5.62 (1.43) 2.53 (0.89) PHQ-9 6.00 (0.95) 6.31 (1.74) 8.18 (1.79) PSYRATS hallucinations 9.15 (2.60) 16.31 (2.84) 1 7.35 (2.23) WHODAS 20.00 (1.80) 23.92 (2.91) 23.06 (2.14) Open in a new tab 1. F = 2.92, df 2, 47, P < .065; AMPH > MEM: P <0.025 (Fisher’s PLSD); AMPH > PBO: d = 0.655. Primary Neurocognitive Outcome MCCB Composite T-scores increased across the course of 30 AT training sessions, independent of drug group. LME analyses of Composite T-scores for T 0 vs P30 revealed a significant main effect of time (T 0 vs P30: b = 3.776, df 48.272, t = 2.494, P = .016), non-significant main effects of AMPH and MEM ( b = 1.854, df 63.384, t = 0.377, P = .707 and b = −0.369, df 63.384, t = −0.081, p = .936, respectively), and non-significant drug-by-time interactions ( b = −3.123, df 48.330, t = −1.295, P = .201 and b = 1.922, df 48.330, t = 0.861, P = .394, respectively). Sensitivity analyses that included all treatment time points (T 0 , P10, P20, and P30) similarly found a significant main effect of time ( P = .01), but non-significant drug main effects and drug-by-time interactions (all P ’s > .75). In PBO group subjects, T 0 mean Composite T-scores (mean (SEM) = 29.85 (3.20)) were > 2 SD below standardized “norms”; mean PBO Composite T-scores peaked at P20 (34.90; mean T-score gains = 5.05), consistent with neurocognitive gains associated with AT ( P < .002). Similar gains were evident in AMPH and MEM group subjects (peak mean T-score gains = 4.54 (P20) and 5.29 (P30) for AMPH and MEM groups, respectively), in whom these gains were maintained at least 12 weeks after P30 ( Table S9 ). Clinical Outcomes Between- and within-group effect sizes for all outcome measures are seen in Tables 3 and 4 (positive between-group d indicates the amount by which clinical or functional improvement over time in an active drug group exceeds that exhibited by PBO group subjects; positive within-group d indicates clinical or functional improvement over time) ( Figure 3A-G ):. Table 3. Outcome Measures: Between-Group Difference (d) 1 in Change from T 1 to P30 Measure AMPH ( n = 13) vs PBO ( n = 20) MEM ( n = 17) vs PBO ( n =2 0) PANSS total 0.364 0.285 PANSS positive 0.688 0.229 PANSS negative −0.037 0.392 YMRS 0.689 −0.081 PHQ-9 0.388 0.399 PSYRATS hallucinations 0.714 0.060 WHODAS 0.696 0.066 Open in a new tab 1. [(mean active drug difference score (T 1 -P30)) – (mean PBO difference score (T 1 -P30)]/SD wav . Bold font = "medium effect size". Table 4. Outcome Measures: Within-Group Change ( d ) 1 from T 1 to Post-Session 30 (P30) Measure PBO ( n = 20) AMPH ( n = 13) MEM ( n = 17) PANSS total 0.355 0.927 2 0.638 PANSS positive 0.230 0.927 2 0.404 PANSS negative 0.172 0.255 0.696 2 YMRS 0.094 0.808 2 0.030 PHQ-9 0.057 0.457 2 0.372 PSYRATS hallucinations −0.072 0.552 2 −0.031 WHODAS −0.055 0.658 2 −0.006 Open in a new tab 1. d = [(score T 1 ) −(score P30)] / SD av (see Methods); positive value = improvement. 2. Meets or exceeds threshold of d > d (PBO) + 0.4. Figure 3. Open in a new tab Outcome measures in PBO group, AMPH group, and MEM group subjects, measured at T 1 , P10, P20, P30, and 12 weeks after completion of AT (“durability assessment” (DA)). Data shown reflects all subjects who completed at least T 1 and P10 testing (T1: n = 60; P10: n = 60; P20: n = 54; P30: n = 50; DA: n = 46). Line interruptions in the X-axis between “P30 and “DA” reflect the transition to the durability assessment phase. For PANSS measures (A-D), declines in AMPH and MEM group subjects during active treatment (from T 1 to P30) were generally steady over time, reaching the “threshold” ( d = 0.4 above PBO value) at P30 among AMPH group subjects for PANSS total (A) and positive subscales (C) (*) and among MEM group subjects for the PANSS negative subscale (D) (#). LME models revealed that reductions in PANSS total and positive scores from T 1 to P30 reached significance in AMPH ( P = .004 and P < .005, respectively) and MEM subjects ( P < .001 and P < .020, respectively) but not in PBO subjects. For PANSS negative scores, there were significant declines in MEM subjects ( P < .013) but not AMPH or PBO subjects. PANSS general subscale scores corresponded closely to PANSS total scores. DA scores for most measures suggested stable or enhanced gains in AMPH group subjects, 12-weeks after cessation of treatment (+; Tables 5 and S9 ). For secondary clinical outcome measures (E-G) and the primary functional outcome measure (H), declines in AMPH group scores to “threshold” levels (*) were detected as early as P10 in YMRS (E), PHQ-9 (F), PSYRATS hallucinations subscale (G) and WHODAS (H), and were maintained through P30. YMRS scores declined over time among AMPH group subjects ( P < .015), reaching threshold by P10 ( d (AMPH) = 0.700 vs d (PBO) = 0.059) and remaining suprathreshold at P20 ( d (AMPH) = 0.518 vs d (PBO) = (−0.088)) and P30 ( d (AMPH) = 0.808 vs d (PBO) = 0.094). A drug effect on the YMRS in MEM subjects might have been obscured by low baseline scores in this group vs AMPH group and PBO group subjects. A decline in PHQ-9 scores reached threshold among AMPH group subjects at P10 ( d (AMPH) = 0.726 vs d (PBO) = 0.025). The PSYRATS hallucination subscale detected elevated baseline scores among AMPH group subjects compared to either MEM group or PBO group subjects ( Table 2 ); scores in AMPH group subjects declined from this elevated baseline, reaching threshold at P10 (d(AMPH) = 0.498 vs d (PBO) = (−0.144)) as well as P30 ( d (AMPH) = 0.552 vs d (PBO) = (−0.072)). The reduction in WHODAS scores among AMPH group subjects reached threshold by P10 (d(AMPH) = 0.264 vs d(PBO) = (−0.268)), in part due to a small increase in WHODAS scores among PBO group subjects. As with the primary outcome measures (A-D), DA scores shown for most secondary measures suggest stable or enhanced gains in AMPH group subjects, 12-weeks after cessation of treatment (+: T 1 vs DA reaches “threshold”) (see Tables 5 and S8 ). LME analyses of the primary clinical outcome measure—PANSS total score—revealed non-significant main effects of AMPH and MEM ( b = 0.767, df 104.086, t = 0.164, P = .870 and b = −0.365, df 104.035, t = −0.084, P = .933, respectively), a significant main effect of time (T 0 vs P30: b = −1.620, df 48.938, t = −2.194, P = .033), and non-significant drug-by-time interactions ( b = −1.582, df 49.088, t = −1.346, P = .185 and b = −1.310, df 48.950, t = −1.202, P = .235, respectively). Between-group d ( Table 3 ) were 0.364 (AMPH vs PBO) and 0.285 (MEM vs PBO). LME within-group models revealed significant declines in PANSS total scores from baseline to P30 in both AMPH ( b = −3.231, df 11.141, t = −3.595, P = .004) and MEM groups ( b = −2.87, df 16.153, t = −4.035, P < .001) and a non-significant trend in the PBO group ( b = −1.603, df 21.347, t = -1.99, P < .06). Within-group d are seen in Table 4 : MEM subjects ( d = 0.638) missed the “threshold” of a d = 0.4 decline in PANSS total score over PBO (i.e., threshold d ≥ 0.355 + 0.4 = 0.755), while AMPH subjects ( d = 0.927) exceeded this threshold. In analyses of secondary clinical measures, LME models detected significant drug-by-time interactions indicating superiority of AMPH vs PBO for PANSS positive ( b = −0.9, df 48.635, t = -2.227, P < .031), YMRS ( b = −0.998, df 48.374, t = −2.286, P < .027) and PSYRATS ( b = −2.2, df 49.838, t = −2.016, P = .049) scores. While these effects did not survive more restrictive alpha levels (0.01), between-group d (AMPH vs PBO) for these measures were 0.688, 0.689, and 0.714, respectively, indicating medium effect sizes. LME within-group models detected significant effects for PANSS positive ( b = −1.128, df 12.974, t = −3.39, P < .005) and YMRS scores ( b = −1.06, df 13.376, t = −2.841, P < .014) in AMPH group subjects, and significant effects for PANSS positive ( b = −0.656, df 15.825, t = −2.598, P < .020) and PANSS negative scores ( b = −0.97, df 18.572, t = −2.756, P < .013) in MEM group subjects; of these effects, only those of PANSS positive scores in AMPH group subjects survived more restrictive alpha levels. PBO group subjects did not exhibit significant declines in any secondary measures. Within-group d ( Table 4 ) exceeded “threshold” values for AMPH group (PANSS positive scores: 0.927; YMRS: 0.808; PHQ-9: 0.457; PSYRATS: 0.658) and for the MEM group (PANSS negative scores: 0.696). Though not an outcome measure, PANSS general scores declined in all groups from baseline to P30 ( d = 0.489, 0.501, and 0.830, for PBO, MEM, and AMPH groups, respectively). Across PANSS measures, patients with the highest baseline symptoms showed the most clinical improvement ( Figure S1A ). Functional Outcome LME analyses of the primary functional outcome measure—WHODAS disability scores ( Figure 3H )—revealed non-significant main effects of AMPH and MEM ( b = 4.952, df 111.987, t = 1.646, P = .103 and b = 3.053, df 111.985, t = 1.089, P = .279, respectively), a non-significant main effect of time (T 0 vs P30: b = 0.089, df 55.111, t = 0.158, P = .875), and significant drug-by-time interaction effects for AMPH ( b = −1.804, df 55.359, t = −2.003, P = .050, indicating marginal superiority of AMPH vs PBO for decreasing WHODAS scores) but not for MEM ( b = −0.113, df 55.131, t = −0.135, P = .893). The between-group d for AMPH vs PBO was 0.696. LME within-group models showed non-significant effects for PBO, AMPH, and MEM groups (PBO: b = 0.115, df 20.384, t = 0.267, P = .792; AMPH: b = −1.647, df 29.000, t = −1.665, P = .107; MEM: b = −0.005, df 18.009, t = −0.010, P = .992). The within-group d (0.658) for WHODAS in AMPH group subjects exceeded the threshold level ( Table 4 ). Non-scored WHODAS “Effect of Disability” items confirmed robust declines among AMPH subjects in the number of days “totally unable to carry out activities” ( d = 1.118) or “needed to reduce activities” ( d = 0.888) due to their illness. Predictors of Clinical Outcome Measures A comprehensive analysis of factors predicting the most robust PACT effects on clinical and functional measures will be included in a separate report on potential biomarker predictors (in preparation). However, two such factors are briefly noted here and in Supplemental Results. First, clinical gains in all groups (AMPH, MEM, and PBO) were most robust among subjects with the highest levels of initial (T 1 ) symptoms ( Figure S1A ). Second, clinical gains in AMPH group subjects tended to be inversely related to AP dosing (CPZ equivalents). In other words, AMPH tended to be least effective in enhancing AT effects on clinical symptoms in subjects with the highest level of dopamine receptor blockade ( Figure S1B ); this was not the case with gains in either MEM or PBO groups. Durability Assessment Forty-six subjects completed 30 AT sessions and P30 testing and returned for “durability” testing 12 weeks later ( Table 5 ). Sample attrition by the time of the durability assessment was 29.6%, 30.4%, and 38.9% for PBO, MEM, and AMPH groups, respectively. Nonetheless, a pattern of results was evident across many outcome measures: 12 weeks after the P30 assessment, within-group measures in AMPH subjects showed stable improvement or further modest gains, while within-group measures in MEM and PBO subjects generally returned to slightly higher (more symptomatic) levels ( Table 5 ; Figure 3 ). Table 5. Outcome Measures: Change (mean (SEM)) from Post-Session 30 to Durability Assessment 1 Measure PBO ( n = 19) AMPH ( n = 11) MEM ( n = 16) PANSS total 2.84 (3.19) 3.36 (2.86) −4.19 (2.45) PANSS positive −0.11 (1.04) 0.09 (0.83) −0.56 (0.93) PANSS negative 1.84 (0.86) 1.18 (0.90) −1.81 (0.62) YMRS −0.42 (0.79) 0.46 (0.31) −2.44 (1.22) PHQ-9 −0.63 (1.27) 0.09 (1.33) −0.81 (1.10) PSYRATS hallucinations −0.21 (2.46) 0.18 (3.62) 2.67 (1.49) WHODAS −0.84 (1.64) 0.36 (1.57) −0.19 (1.44) Open in a new tab 1. Durability Assessment (DA) completed 12 weeks after last TCT session. Values represent a “difference score”: P30 minus DA (positive value = continued improvement). To establish the “durability” of changes in outcome measures, we calculated the effect size changes from T1 to durability assessment (DA) (i.e., d = (score at T 1 —score at DA )/SD av ) and generated “threshold” values for active drug groups of Cohen’s d (active drug) ≥ Cohen’s d (PBO) + 0.4). As seen in Figure 3 and Tables S8 and S9 , after a 12-week “break” from AT and pharmacologic augmentation, AMPH group subjects sustained “suprathreshold” gains in most measures. AMPH and MEM group subjects also reached peak neurocognitive performance at the DA, when mean MCCB Composite T-score gains from T 0 were 2.80, 6.93, and 5.56 for PBO, AMPH, and MEM groups, respectively (Table S9 ). Treatment Satisfaction After 30 sessions of AT, subjects in all groups were more satisfied and felt the treatment was more worthwhile than they had estimated prior to initiating AT. Scores for expected vs actual “treatment satisfaction” did not differ across drug groups; scores increased from T 0 (prior to the first AT training session, which reflects “expected” satisfaction) to post-session 30 (which reflects “actual” satisfaction; F = 15.976, df 1,47, P < .0005) with no significant effect of drug or drug × time interaction (both F ’s < 1). Ratings of expected vs actual “hard work” showed no significant effect of time ( F < 1) or drug ( F = 2.226, df 2,47, ns), but there was a significant time × drug interaction ( F = 4.336, df 2,47, P < .019). This interaction reflected the fact that AMPH group subjects tended to experience the treatment as “less hard work” than they expected ( d = 0.448), while MEM subjects tended to experience the treatment as “more hard work” than they expected ( d = 0.538). Overall ratings of expected vs actual “worthwhile” increased from T 0 to post-session 30 ( F = 9.432, df 1,47, P < .004), with no significant effect of drug or time × drug interaction (both F ’s < 1). Because these “expected vs. actual” comparisons are limited to subjects who completed at least post-session 30, they are conceivably biased towards more favorable assessments (i.e., if less-satisfied subjects were more likely to withdraw from the study and thereby not register post-session 30 values). However, “expected” satisfaction ( F < 1), hard work ( F = 1.154, df 1,59, ns) and worth ( F < 1) did not differ significantly between subjects who did vs did not complete the full course of AT. Discussion This study tested the hypothesis that, in AP-medicated individuals with psychosis, the beneficial neurocognitive, clinical, and functional effects of AT can be augmented by drugs previously found to enhance AT learning. Participants completed up to 30 h of AT. Clinical effects of 3 regimens—AT + PBO, AT + AMPH, and AT + MEM—were tested using a broad battery of measures. The Results were Mixed Compared to individuals receiving AT + PBO, gains in the primary clinical (PANSS total score) and neurocognitive measures (MCCB Composite Score) were not statistically greater among AT + AMPH or AT + MEM group subjects. On the other hand, gains in the primary functional outcome measure (WHODAS) were marginally significantly greater among AT + AMPH vs AT + PBO subjects ( P = .050). Moreover, 3 of 5 secondary outcome measures detected “superiority” of AT + AMPH vs AT + PBO (PANSS positive, YMRS, PSYRATS), albeit at levels ( P ’s ≤ .027–.049) that would not survive conservative corrections for multiple comparisons. And while within-group analyses detected no significant clinical gains from baseline to P30 among AT + PBO group subjects, such gains were detected in primary and secondary clinical measures in AT + AMPH and AT + MEM groups; three of these gains survived corrections for multiple comparisons (PANSS total (AT + AMPH and AT + MEM), PANSS positive (AT + AMPH); P ’s ≤ .001–.005), while 3 others did not (PANSS positive and PANSS negative (AT + MEM), YMRS (AT + AMPH); P ’s ≤ .013–.020). Lastly, “suprathreshold” improvement (exceeding AT + PBO group gains by at least d = 0.4) was detected among AT + AMPH subjects in measures of PANSS total and positive scores, YMRS, PHQ-9, PSYRATS, and WHODAS. AT + MEM group subjects experienced suprathreshold improvement in PANSS negative scores, though AT + MEM-associated benefits also exceeded those in AT + PBO group subjects by non-trivial margins for PANSS total (by d = 0.283) and PHQ-9 scores (by d = 0.315). By and large, gains associated with AMPH were “durable,” i.e., persisted or increased over the 12 weeks after the end of active treatment, while gains among AT + MEM and AT + PBO groups diminished during the 12-week post-treatment phase. In contrast to gains in the primary functional and secondary clinical measures, neurocognitive gains across the training intervals were independent of drug type (PBO, AMPH, or MEM). With no AT “control” group, this study was not optimally designed to assess the neurocognitive, clinical, or functional benefits of AT alone. Nonetheless, consistent with the published literature, AT + PBO group subjects exhibited significant neurocognitive gains over time (MCCB composite; P < .002), as well as trends towards within-subject clinical gains (PANSS total score: d = 0.355; P < .060). The active drug regimens in this study were very distinct. AMPH group subjects were exposed to a low dose (5 mg) of short-acting AMPH ( t ½ ≈10 h), 2-3 times per week; 5 mg is a pediatric starting dose of AMPH for Attention Deficit Hyperactivity Disorder. 68 Clinical and functional gains in AMPH group subjects were detected on days when they were not receiving AMPH. Thus, based on the low dose, its short half-life, and the fact that measures were assessed on “non-drug days,” it seems unlikely that the gains in AMPH group subjects reflect its acute neurochemical effects. Daily administration of higher effective doses of an AMPH analog to AP-treated SZ patients, without AT, is associated with reduced negative symptoms 69 ; this was not detected with the current AT + AMPH regimen. In contrast, MEM was titrated to a daily dose (10 mg twice daily) maintained throughout the course of the 30 AT sessions and the P30 testing. All clinical assessments during P10, P20, and P30 were presumably conducted in the context of robust steady-state levels of MEM. This regimen of MEM added to APs is reported in meta-analyses to reduce negative—and in some cases positive—symptoms in SZ. 70–74 Thus, reductions in PANSS scores in MEM group subjects may have reflected the salutary effects of MEM, independent of any gains promoted by AT. Conceivably, benefits seen here with AMPH might reflect its pro-attentional and wakefulness-promoting properties, helping AP-medicated patients stay awake and attentive during 60-90 minutes of repetitive AT tasks. Interestingly, the wakefulness-promoter modafinil (200 mg) added to 10 AT sessions did not improve PANSS scores or neurocognition in SZ patients. 75 Perhaps this was not an adequate test: in the present study, “suprathreshold” reductions in PANSS scores with AMPH required 30 AT sessions. Targeting psychostimulants as potential treatments for psychosis is clearly problematic: many psychostimulants carry liability for abuse, and in some cases for exacerbating psychosis in SZ. The latter effect is challenged by findings of reduced hospitalizations among over 2200 AP-medicated psychosis patients within the 12 months after starting treatment with a psychostimulant including AMPH. 76 Importantly, the fact that this PACT model may involve staff-monitored pill administration, at a limited frequency (prior to each of 30 AT sessions) and low dose, may mitigate against some risks associated with the use of psychostimulants to augment AT effects. Cain et al. 77 reported a dissociation of gains in clinical symptoms vs neurocognition in PACT trials that combined D-cycloserine (DCS) with AT in SZ patients; in that study, adding DCS to AT enhanced reductions of negative symptoms, while neurocognitive gains were only evident in the PBO + AT group. In the present study, both PBO and active drug groups exhibited gains in MCCB composite scores; peak gains largely exceeded levels associated with “practice effects.” 65 Unlike their effects on several clinical measures and the one functional measure, neither AMPH nor MEM augmented AT-associated gains in MCCB performance when compared to PBO, from baseline to P30 values. Interestingly, pharmacologic augmentation of neurocognitive gains appears more robust when compared from baseline to DA (T score gains = 2.80, 6.93, and 5.56 for PBO, AMPH, and MEM groups, respectively). Loewy et al. 20 also reported the extension of AT-associated neurocognitive gains during a post-training hiatus. More generally, our findings suggest at least a temporal dissociation between drug effects on AT-enhanced clinical/functional vs. neurocognitive measures (e.g., in AT + AMPH subjects at P30, maximal reductions in PANSS total scores ( d = 0.927) coincide with minimal gains in MCCB composite scores (T-score gain = 0.77)). Conceivably, this dissociation may reflect differential “practice effects” on clinical/functional vs neurocognitive measures, or different “ceiling” levels between these measures, i.e., there may be more restrictive limits to the amount by which neurocognition can be improved through the addition of drugs to AT. Perhaps most importantly, the present findings suggest that clinical gains imparted by AT + active drugs are not causally linked to drug-enhanced neurocognition (consistent with Cain et al. 77 ). Limitations These findings must be interpreted in the context of significant limitations: No AT “Control”: AT + PBO group subjects exhibited non-trivial reductions in PANSS total ( d = 0.355) and PANSS general subscale scores ( d = 0.489); with no AT “control,” we do not know whether some or all of these gains were independent of AT, perhaps reflecting the impact of the laboratory’s structured and supportive “day treatment” milieu. It is likely that subjects benefitted from study participation, based on diverse factors including increased socialization, circadian entrainment, monetary gains, and greater sense of purpose and self-esteem. Small Sample: Only 13 AMPH group subjects and 17 MEM group subjects completed the 30-session course of AT in this study; the stability of these findings should be tested in larger, multi-site cohorts. In addition to the pandemic, which precluded or restricted subject recruitment for more than 1.5y, the small sample of AMPH group subjects reflected factors more generally relevant to any future use of psychostimulants to augment AT effects. First, there is a high comorbidity of psychosis and stimulant use disorders 78 which was exclusionary and thus limited enrollment. Second, structured living settings and community clinicians are reluctant to permit/refer their psychosis patients to join a trial involving AMPH. Despite “real-world” data to the contrary, 76 there is 60 years of literature linking AMPH to the exacerbation of psychosis in SZ (but see 79–81 ). While the biology of AMPH effects in the present PACT model (5 mg, 2-3 times/week, in AP-medicated patients) differs categorically from the biology of high-dose, escalating AMPH regimens known to be psychotogenic, the suggestion of using AMPH as part of a therapeutic model for SZ meets with substantial resistance. Presumably, both of these factors—comorbid stimulant misuse, and resistance of clinicians to engage patients in AMPH-related treatment—might also limit future applications of this PACT model and underscores the need to identify drugs other than psychostimulants as potential AT-enhancing agents. Arbitrary “Threshold”: Data were analyzed using both between- and within-group LME models, as well as between- and within-group effect sizes. To interpret the relevance of the within-group effect sizes, we set a value of “ d ≥ 0.4 above PBO group” as the threshold for achieving meaningful benefit with either AMPH or MEM. This “threshold” is somewhat arbitrary and potentially restrictive. We anticipated AT + PBO group clinical effects in the range of d = 0.4 17 , 18 ; a threshold set higher than d = 0.4 would require the AT + active drug intervention to yield large effect size benefits ( d ≥ (0.4 + 0.4) or d ≥ 0.8). Thus, in choosing “ d ≥ 0.4 above PBO group,” we applied a conservative threshold with a low likelihood of false-positive conclusions. For reference, among the primary outcome measures, actual AT + PBO group effect sizes ranged from d = 0.172 (PANSS negative score) to d = 0.355 (PANSS total score). Based on these values, an added d = 0.4 reduces the “number needed to treat” (NNT) from 5.05 to 2.46 for PANSS total and from 10.33 to 3.18 for PANSS negative scores (see Table S7 for actual NNT values 82 ). Increasing the threshold from d = 0.4 to d = 0.5 (“medium effect size”) excludes only 1 secondary measure (AMPH effect on PHQ-9; Table S6 ), while decreasing this threshold to d = 0.3 adds only 1 measure (MEM effect on PHQ-9; Table S6 ). Other threshold definitions (e.g., doubling PBO group gains) yield similar findings. We originally proposed to use “traditional” tests of statistical significance to assess treatment outcomes in separate studies of AMPH- and MEM-augmented AT; the originally proposed sample sizes provided robust power to detect statistically significant changes in all primary outcome measures. As noted above, we were unable to achieve the proposed sample sizes due to pandemic-related limitations on recruitment and testing. Post-hoc steps taken to address this dilemma included combining the 2 separate studies, “pooling” PBO groups, and using between- and within-group effect sizes, in addition to LME models, to assess changes in outcome measures. This use of effect sizes as a metric of treatment effect is broadly supported in the literature, 82 especially when sample sizes are small. In fact, it is argued that “traditional” tests often detect robust statistical significance despite small (and clinically inconsequential) group differences, based on power provided by large sample sizes. 62 Suboptimal AT: AT is most effective if delivered within an integrated treatment program, in conjunction with evidence-based psychotherapies, social skills training, nutritional consultation, physical exercise 19 , 83 , 84 and medication management that minimizes sedative effects and anticholinergic burden 85 , 86 during AT training. Even simple steps such as providing feedback to subjects regarding their AT performance, or periodically engaging and reorienting subjects to the goals of AT exercises, might boost the real-life clinical impact of AT (e.g. 83 ). These interventions were not used here, to avoid confounding the test of the PACT model. MEM Adherence: AMPH adherence was confirmed on-site; MEM was self-administered or distributed by staff at assisted living centers, hence evidence of adherence was indirect. Patients and staff were queried and generally confirmed proper dosing (see Supplement, “Attrition”: 3 missed doses out of 1013 queries). While the long half-life of MEM 87 compensates for some consequences of missed doses, it is a limitation that MEM blood levels were not acquired to confirm adherence. Small Gains: The absolute magnitude of gains in active drug groups was modest. For example, the large effect size within-subject improvement in PANSS total score among AT + AMPH group subjects ( d = 0.927) reflects a mean decline of 9.077 points, a 16.32% reduction from a T 1 average of 55.615 to a P30 average of 46.438. A “clinically meaningful” reduction in PANSS total score is suggested to be 34%, or 18.909 points from the T 1 value. 88 Applying this criterion, even large effect size PANSS reductions detected in active drug groups may not be clinically meaningful (but see Table S7 for NNTs). For some measures (e.g., PANSS total score; Figure 3A ), it appears that AMPH- and MEM-induced gains might have increased with longer AT training. Multiple Comparisons: The numbers of primary (3) and secondary (5) outcome measures is modest, and most were selected as a priori predictions based on published findings (including from our own group) of AT effects in SZ cohorts. Nonetheless, using a conservative analytic approach that corrects alpha for these multiple comparisons, most of the reported findings would no longer achieve statistical significance. Those surviving these more rigorous significance levels would include within-subject effects of AMPH ( P = .004) and MEM ( P < .001) for the PANSS total score, and within-subject effects of AMPH ( P < .005) for the PANSS positive score. Conclusions Seven decades after the introduction of APs, a treatment model that effectively controls the constellation of symptoms and functional impairment in psychosis remains elusive. APs do not fundamentally change the most disabling effects of SZ and are associated with health-disruptive effects that contribute to a life expectancy shortened by 15 years. 89 Using AT to engage the brain’s mechanisms of learning and neuroplasticity is a paradigm shift that holds promise for enhancing cognitive, clinical and functional rehabilitation in psychosis. To address the modest and variable gains associated with AT, we proposed using medications to enhance AT learning by bolstering basic mechanisms—like attention and auditory processing—critical to the therapeutic effects of AT. Preliminary results of this “PACT” strategy were mixed. Gains in the primary clinical and neurocognitive outcome measures among active drug group subjects did not significantly exceed those among PBO group subjects. However, significant gains in the primary functional measure and several secondary clinical measures—above those detected in AT + PBO group subjects—were evident when AMPH and to a lesser degree MEM was paired with AT. Within-subjects, pairing these active drugs with AT was associated with medium-to-large effect size benefits in psychosis patients across several clinical and functional measures; such gains were not seen when PBO was added to AT. Both AMPH and MEM were well-tolerated, and patients expressed satisfaction with these PACT treatments. There are cultural and logistical disadvantages to the use of these medications—particularly AMPH—in chronic psychosis, that might be countered by confirmatory safety and outcome data from larger cohorts. PACT models for psychosis remain worthwhile targets for investigation. Supplementary Material sbaf015_suppl_Supplementary_Tables_S1-S9_Figures_S1-S9 sbaf015_suppl_supplementary_tables_s1-s9_figures_s1-s9.docx (335.6KB, docx) Acknowledgments The authors acknowledge the excellent technical assistance of Lindsay Wasserman-Hallock, Dylan Iwanaga, Ana Zamudio Aispuro, Melina Garay, Connor Warman, Laura MacDonald, Abigail Potter and Addison Denning and the exceptional administrative support of Ms Maria Bongiovanni. Contributor Information Neal R Swerdlow, Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States; VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States. Joyce Sprock, Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States; VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States. Francesca Li, Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States. Jenny Min Din, Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States. Jessica Minhas, Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States. Jo Talledo, Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States. Yash B Joshi, Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States; VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States. Juan L Molina, Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States; VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States. Bethany Nordberg, Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States. Kevin Ing, Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States; VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States. Michael L Thomas, Department of Psychology, Colorado State University, Fort Collins, CO 80523, United States. Gregory A Light, Department of Psychiatry, UCSD School of Medicine, La Jolla, CA 92093, United States; VISN-22 Mental Illness Research Education and Clinical Center, VA San Diego Healthcare System, La Jolla, CA 92037, United States. Author Contributions F. Li, J. Minhas, J. Min Din, J. Sprock, J. Talledo, and B. Norberg collected the data. N. Swerdlow, M. Thomas, and G. Light analyzed the data. N. Swerdlow, G. Light, J. Molina, M. Thomas, and Y. Joshi interpreted the results. N. Swerdlow and G. Light designed the study and supervised all aspects of collection, analysis, and interpretation of the data. N. Swerdlow wrote the original manuscript; edits were solicited from all authors. All authors contributed to and approved the final manuscript. Funding This study was supported by National Institute of Mental Health (R33 MH123603 to NRS and R33 MH125114 to GAL) and the Veterans Affairs VISN-22 Mental Illness Research, Education, and Clinical Center (MIRECC). GAL is also supported by an award from the Sidney R. Baer, Jr Foundation. YBJ is supported by the Veterans Affairs CDA2 grant 1IK2RX003395; JLM is supported by Veterans Affairs CDA2 grant IK2 RX004570 and by a Brain and Behavior Research Foundation Young Investigator Award #28927. Conflicts of Interest GAL is a consultant for Johnson & Johnson, Neurocrine, NeuroSig, and Sosei-Heptares. All other authors have no actual or potential conflicts to report, financial or otherwise. Data Availability Raw T1 and P30 data for all PANSS measures among “completers” (n = 50) used to calculate within-subject effect sizes are found in Supplemental Results ( Table S5 ); remaining data is available on request, pending publication of findings. References 1. Bowie  CR, Leung  WW, Reichenberg  A, et al.  Predicting schizophrenia patients’ real-world behavior with specific neuropsychological and functional capacity measures. Biol Psychiatry.  2008;63:505–511. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 2. Green  MF.  What are the functional consequences of neurocognitive deficits in schizophrenia? Am J Psychiatry.  1996;153:321–330. [ DOI ] [ PubMed ] [ Google Scholar ] 3. Wykes  T.  Predicting symptomatic and behavioural outcomes of community care. Br J Psychiatry.  1994;165:486–492. [ DOI ] [ PubMed ] [ Google Scholar ] 4. Keefe  RS, Bilder  RM, Davis  SM, et al. ; CATIE Investigators. Neurocognitive effects of antipsychotic medications in patients with chronic schizophrenia in the CATIE Trial. Arch Gen Psychiatry.  2007;64:633–647. [ DOI ] [ PubMed ] [ Google Scholar ] 5. Barretto  EM, Kayo  M, Avrichir  BS, et al.  A preliminary controlled trial of cognitive behavioral therapy in clozapine resistant schizophrenia. J Nerv Ment Dis.  2009;197:865–868. [ DOI ] [ PubMed ] [ Google Scholar ] 6. Eack  SM, Hogarty  GE, Cho  RY, et al.  Neuroprotective effects of cognitive enhancement therapy against gray matter loss in early schizophrenia: results from a 2-year randomized controlled trial. Arch Gen Psychiatry.  2010;67:674–682. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. Granholm  E, McQuaid  JR, McClure  FS, et al.  Randomized controlled trial of cognitive behavioral social skills training for older people with schizophrenia: 12-month follow-up. J Clin Psychiatry.  2007;68:730–737. [ DOI ] [ PubMed ] [ Google Scholar ] 8. Twamley  EW, Narvaez  JM, Becker  DR, Bartels  SJ, Jeste  DV.  Supported employment for middle-aged and older people with schizophrenia. Am J Psychiatr Rehabil. 2008a;11:76–89. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Twamley  EW, Savla  GN, Zurhellen  CH, Heaton  RK, Jeste  DV.  Development and pilot testing of a novel compensatory cognitive training intervention for people with psychosis. Am J Psychiatr Rehabil.  2008b;11:144–163. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 10. McGurk  SR, Mueser  KT, DeRosa  TJ, Wolfe  R.  Work, recovery, and comorbidity in schizophrenia: a randomized controlled trial of cognitive remediation. Schizophr Bull.  2009;35:319–335. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 11. Vita  A, Barlati  S, Ceraso  A, et al.  Durability of effects of cognitive remediation on cognition and psychosocial functioning in schizophrenia: A systematic review and meta-analysis of randomized clinical trials. Am J Psychiatry.  2024;181:520–531. [ DOI ] [ PubMed ] [ Google Scholar ] 12. Kurtz  MM.  Cognitive remediation in schizophrenia spectrum illness: Evidence for treatment persistence. Am J Psychiatry.  2024;181:471–473. [ DOI ] [ PubMed ] [ Google Scholar ] 13. McGurk  SR, Twamley  EW, Sitzer  DI, McHugo  GJ, Mueser  KT.  A meta-analysis of cognitive remediation in schizophrenia. Am J Psychiatry.  2007;164:1791–1802. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Fisher  M, Holland  C, Merzenich  MM, Vinogradov  S.  Using neuroplasticity-based auditory training to improve verbal memory in schizophrenia. Am J Psychiatry.  2009;166:805–811. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 15. Fisher  M, Holland  C, Subramaniam  K, Vinogradov  S.  Neuroplasticity-based cognitive training in schizophrenia: an interim report on the effects 6 months later. Schizophr Bull.  2010;36:869–879. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 16. Vinogradov  S, Fisher  M, de Villers-Sidani  E.  Cognitive training for impaired neural systems in neuropsychiatric illness. Neuropsychopharmacology.  2012;37:43–76. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 17. Thomas  ML, Bismark  AW, Joshi  YB, et al.  Targeted cognitive training improves auditory and verbal outcomes among treatment refractory schizophrenia patients mandated to residential care. Schizophr Res.  2018a;202:378–384. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Thomas  ML, Treichler  EBH, Bismark  A, et al.  Computerized cognitive training is associated with improved psychosocial treatment engagement in schizophrenia. Schizophr Res.  2018b;202:341–346. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Miley  K, Fisher  M, Nahum  M, et al.  Six month durability of targeted cognitive training supplemented with social cognition exercises in schizophrenia. Schizophr Res Cogn. 2019;20:100171. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 20. Loewy  R, Fisher  M, Ma  S, et al.  Durable cognitive gains and symptom improvement are observed in individuals with recent-onset schizophrenia 6 months after a randomized trial of auditory training completed remotely. Schizophr Bull.  2022;48:262–272. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Keefe  RS, Vinogradov  S, Medalia  A, et al.  Feasibility and pilot efficacy results from the multisite Cognitive Remediation in the Schizophrenia Trials Network (CRSTN) randomized controlled trial. J Clin Psychiatry.  2012;73:1016–1022. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Murthy  NV, Mahncke  H, Wexler  BE, et al.  Computerized cognitive remediation training for schizophrenia: an open label, multi-site, multinational methodology study. Schizophr Res.  2012;139:87–91. [ DOI ] [ PubMed ] [ Google Scholar ] 23. Fisher  M, Loewy  R, Carter  C, et al.  Neuroplasticity-based auditory training via laptop computer improves cognition in young individuals with recent onset schizophrenia. Schizophr Bull.  2015;41:250–258. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Scoriels  L, Genaro  LT, Mororó  LGC, et al.  Auditory versus visual neuroscience-informed cognitive training in schizophrenia: Effects on cognition, symptoms and quality of life. Schizophr Res.  2020;222:319–326. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 25. Loewy  R, Fisher  M, Schlosser  DA, et al.  Intensive auditory cognitive training improves verbal memory in adolescents and young adults at clinical high risk for psychosis. Schizophr Bull.  2016;42:S118–S126. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 26. Subramaniam  K, Luks  TL, Garrett  C, et al.  Intensive cognitive training in schizophrenia enhances working memory and associated prefrontal cortical efficiency in a manner that drives long-term functional gains. Neuroimage.  2014;99:281–292. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Prikken  M, Konings  MJ, Lei  WU, Begemann  MJH, Sommer  IEC.  The efficacy of computerized cognitive drill and practice training for patients with a schizophrenia-spectrum disorder: A meta-analysis. Schizophr Res.  2019;204:368–374. [ DOI ] [ PubMed ] [ Google Scholar ] 28. Swerdlow  NR.  Are we studying and treating schizophrenia correctly? Schizophr Res.  2011a;130:1–10. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Swerdlow  NR.  Beyond antipsychotics: Pharmacologically-augmented cognitive therapies (PACTs) for schizophrenia. Neuropsychopharmacology.  2011b;37:310–311. [ Google Scholar ] 30. Barch  DM.  Pharmacological strategies for enhancing cognition in schizophrenia. Curr Top Behav Neurosci. 2010;4:43–96. [ DOI ] [ PubMed ] [ Google Scholar ] 31. Buchanan  RW, Javitt  DC, Marder  SR, et al.  The Cognitive and Negative Symptoms in Schizophrenia Trial (CONSIST): the efficacy of glutamatergic agents for negative symptoms and cognitive impairments. Am J Psychiatry.  2007;164:1593–1602. [ DOI ] [ PubMed ] [ Google Scholar ] 32. Goff  DC, Keefe  R, Citrome  L, et al.  Lamotrigine as add-on therapy in schizophrenia: results of 2 placebo-controlled trials. J Clin Psychopharmacol.  2007;27:582–589. [ DOI ] [ PubMed ] [ Google Scholar ] 33. Goff  DC, Tsai  G, Levitt  J, et al.  A placebo-controlled trial of D-cycloserine added to conventional neuroleptics in patients with schizophrenia. Arch Gen Psychiatry.  1999;56:21–27. [ DOI ] [ PubMed ] [ Google Scholar ] 34. Goff  DC, Tsai  G, Manoach  DS, Flood  J, Darby  DG, Coyle  JT.  D-cycloserine added to clozapine for patients with schizophrenia. Am J Psychiatry.  1996;153:1628–1630. [ DOI ] [ PubMed ] [ Google Scholar ] 35. Green  MF.  Cognition, drug treatment, and functional outcome in schizophrenia: A tale of two transitions. Am J Psychiatry.  2007;164:992–994. [ DOI ] [ PubMed ] [ Google Scholar ] 36. Choi  DC, Rothbaum  BO, Gerardi  M, Ressler  KJ.  Pharmacological enhancement of behavioral therapy: focus on posttraumatic stress disorder. Curr Top Behav Neurosci. 2010;2:279–299. [ DOI ] [ PubMed ] [ Google Scholar ] 37. Ressler  KJ, Rothbaum  BO, Tannenbaum  L, et al.  Cognitive enhancers as adjuncts to psychotherapy: use of D-cycloserine in phobic individuals to facilitate extinction of fear. Arch Gen Psychiatry.  2004;61:1136–1144. [ DOI ] [ PubMed ] [ Google Scholar ] 38. Swerdlow  NR, Tarasenko  M, Bhakta  SG, et al.  Amphetamine enhances gains in auditory discrimination training in adult schizophrenia patients. Schizophr Bull.  2017;43:872–880. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 39. Swerdlow  NR, Bhakta  SG, Talledo  J, et al.  Auditory discrimination and frequency modulation learning in schizophrenia patients: amphetamine within-subject dose response and time course. Psychol Med.  2023;53:140–148. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 40. Swerdlow  NR, Bhakta  SG, Talledo  J, et al.  Memantine effects on auditory discrimination and training in schizophrenia patients. Neuropsychopharmacology.  2020;45:2180–2188. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 41. Sheehan  DV, Lecrubier  Y, Sheehan  KH, et al.  The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry.  1998;59:22–33;quiz 34. [ Google Scholar ] 42. Adcock  RA, Dale  C, Fisher  M, et al.  When top-down meets bottom-up: auditory training enhances verbal memory in schizophrenia. Schizophr Bull.  2009;35:1132–1141. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 43. Biagianti  B, Fisher  M, Neilands  TB, Loewy  R, Vinogradov  S.  Engagement with the auditory processing system during targeted auditory cognitive training mediates changes in cognitive outcomes in individuals with schizophrenia. Neuropsychology.  2016;30:998–1008. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 44. Molina  JL, Joshi  YB, Nungaray  JA, et al.  Central auditory processing deficits in schizophrenia: Effects of auditory-based cognitive training. Schizophr Res.  2021;236:135–141. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 45. Molina  JL, Thomas  ML, Joshi  YB, et al.  Gamma oscillations predict pro-cognitive and clinical response to auditory-based cognitive training in schizophrenia. Transl Psychiatry.  2020;10:405. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 46. Hochberger  WC, Joshi  YB, Thomas  ML, et al.  Neurophysiologic measures of target engagement predict response to auditory-based cognitive training in treatment refractory schizophrenia. Neuropsychopharmacology.  2019a;44:606–612. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 47. Hochberger  WC, Thomas  ML, Joshi  YB, et al.  Oscillatory biomarkers of early auditory information processing predict cognitive gains following targeted cognitive training in schizophrenia patients. Schizophr Res.  2019b;215:97–104. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 48. Joshi  YB, Thomas  ML, Hochberger  WC, et al.  Verbal learning deficits associated with increased anticholinergic burden are attenuated with targeted cognitive training in treatment refractory schizophrenia patients. Schizophr Res.  2019;208:384–389. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 49. Molina  JL, Joshi  YB, Nungaray  JA, et al.  Early auditory processing abnormalities alter individual learning trajectories and sensitivity to computerized cognitive training in schizophrenia. Psychol Med.  2024;54:2669–2676. [ Google Scholar ] 50. Tarasenko  M, Perez  VB, Pianka  ST, et al.  Measuring the capacity for auditory system plasticity: An examination of performance gains during initial exposure to auditory-targeted cognitive training in schizophrenia. Schizophr Res.  2016;172:123–130. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 51. Posner  K, Brown  GK, Stanley  B, et al.  The Columbia-Suicide Severity Rating Scale: initial validity and internal consistency findings from three multisite studies with adolescents and adults. Am J Psychiatry.  2011;168:1266–1277. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 52. Swerdlow  NR, Bhakta  S, Chou  HH, Talledo  JA, Balvaneda  B, Light  GA.  Memantine effects on sensorimotor gating and mismatch negativity in patients with chronic psychosis. Neuropsychopharm  2016;41:419–430. [ Google Scholar ] 53. Light  GA, Zhang  W, Joshi  YB, Bhakta  S, Talledo  J, Swerdlow  NR.  Single-dose memantine improves cortical oscillatory response dynamics in patients with schizophrenia. Neuropsychopharmacology.  2017;42:2633–2639. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 54. Kay  SR, Fiszbein  A, Opler  LA.  The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull.  1987;13:261–276. [ DOI ] [ PubMed ] [ Google Scholar ] 55. Ustun  TB, Chatterji  S, Kostanjsek  N, et al. ; WHO/NIH Joint Project. Developing the World Health Organization Disability Assessment Schedule 2.0. Bull World Health Organ.  2010;88:815–823. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 56. Spitzer  RL, Kroenke  K, Williams  JBW.  Patient Health Questionnaire Study Group. Validity and utility of a self-report version of PRIME-MD: the PHQ Primary Care Study. JAMA.  1999;282:1737–1744. [ DOI ] [ PubMed ] [ Google Scholar ] 57. Young  RC, Biggs  JT, Ziegler  VE, Meyer  DA.  A rating scale for mania: reliability, validity and sensitivity. Br J Psychiatry.  1978;133:429–435. [ DOI ] [ PubMed ] [ Google Scholar ] 58. Haddock  G, McCarron  J, Tarrier  N, Faragher  EB.  Scales to measure dimensions of hallucinations and delusions: the psychotic symptom rating scales (PSYRATS). Psychol Med.  1999;29:879–889. [ DOI ] [ PubMed ] [ Google Scholar ] 59. Pinheiro  JC, Bates  DM.  Linear mixed-effects models: Basic concepts and examples. In: Mixed-Effects Models in S and S-PLUS. Statistics and Computing. Springer, New York (NY); 2000, pp. 3–56. [ Google Scholar ] 60. Cohen  J.  Statistical Power Analysis for the Behavioral Sciences. L. Erlbaum Associates, 2nd ed., Hillsdale (NJ); 1988. [ Google Scholar ] 61. Cumming  G.  Understanding the New Statistics: Effect Sizes, Confidence Intervals, and Meta-Analysis. Routledge: New York; 2012. [ Google Scholar ] 62. Cohen  J. Things I have learned (so far). Am Psychol. 1990;45:1304–1312. [ Google Scholar ] 63. Kraemer  HC.  Reporting the size of effects in research studies to facilitate assessment of practical or clinical significance. Psychoneuroendocrinology.  1992;17:527–536. [ DOI ] [ PubMed ] [ Google Scholar ] 64. Kraemer  HC, Neri  E, Spiegel  D.  Wrangling with p-values versus effect sizes to improve medical decision-making: A tutorial. Int J Eat Disord.  2020;53:302–308. [ DOI ] [ PubMed ] [ Google Scholar ] 65. Keefe  RSE, Davis  VG, Harvey  PD, et al.  Placebo response and practice effects in schizophrenia cognition trials. JAMA Psychiatry. 2017;74:807–814. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 66. Guy  W.  ECDEU Assessment manual for psychopharmacology. Abnormal Involuntary Movement Scale (117-AIMS). National Institute of Mental Health: Rockville (MD); 1996. p. 534–537. [ Google Scholar ] 67. McGregor  C, Srisurapanont  M, Mitchell  A, Longo  MC, Cahill  S, White  JM.  Psychometric evaluation of the amphetamine cessation symptom assessment. J Subst Abuse Treat.  2008;34:443–449. [ DOI ] [ PubMed ] [ Google Scholar ] 68. Shoar  NS, Marwaha  R, Molla  M.  Dextroamphetamine-Amphetamine. StatPearls. StatPearls Publishing LLC: Treasure Island (FL); 2024. [ Google Scholar ] 69. Lasser  RA, Dirks  B, Nasrallah  H, et al.  Adjunctive lisdexamfetamine dimesylate therapy in adult outpatients with predominant negative symptoms of schizophrenia: open-label and randomized-withdrawal phases. Neuropsychopharmacology.  2013;38:2140–2149. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 70. Kishi  T, Matsuda  Y, Iwata  N.  Memantine add-on to antipsychotic treatment for residual negative and cognitive symptoms of schizophrenia: a meta-analysis. Psychopharmacology (Berl).  2017;234:2113–2125. [ DOI ] [ PubMed ] [ Google Scholar ] 71. Etchecopar-Etchart  D, Yon  DK, Wojciechowski  P, et al.  Comprehensive evaluation of 45 augmentation drugs for schizophrenia: a network meta-analysis. EClinicalMedicine. 2024;69:102473. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 72. Vayısoğlu  S, Karahan  S, Anıl Yağcıoğlu  AE.  Augmentation of Antipsychotic Treatment with Memantine in Patients with Schizophrenia: A Systematic Review and Meta-Analysis. Turk Psikiyatri Derg.  2019;30:253–259. [ PubMed ] [ Google Scholar ] 73. Yeh  TC, Correll  CU, Yang  FC, et al.  Pharmacological and nonpharmacological augmentation treatments for clozapine-resistant schizophrenia: A systematic review and network meta-analysis with normalized entropy assessment. Asian J Psychiatr. 2023;79:103375. [ DOI ] [ PubMed ] [ Google Scholar ] 74. Zheng  W, Li  XH, Yang  XH, et al.  Adjunctive memantine for major mental disorders: A systematic review and meta-analysis of randomized double-blind controlled trials. Schizophr Res.  2019;209:12–21. [ DOI ] [ PubMed ] [ Google Scholar ] 75. Michalopoulou  PG, Lewis  SW, Drake  RJ, et al.  Modafinil combined with cognitive training: pharmacological augmentation of cognitive training in schizophrenia. Eur Neuropsychopharmacol.  2015;25:1178–1189. [ DOI ] [ PubMed ] [ Google Scholar ] 76. Corbeil  O, Brodeur  S, Courteau  J, et al.  Treatment with psychostimulants and atomoxetine in people with psychotic disorders: reassessing the risk of clinical deterioration in a real-world setting. Br J Psychiatry.  2024;224:98–105. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 77. Cain  CK, McCue  M, Bello  I, et al.  d-Cycloserine augmentation of cognitive remediation in schizophrenia. Schizophr Res.  2014;153:177–183. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 78. Hunt  GE, Large  MM, Cleary  M, Lai  HMX, Saunders  JB.  Prevalence of comorbid substance use in schizophrenia spectrum disorders in community and clinical settings, 1990-2017: Systematic review and meta-analysis. Drug Alcohol Depend.  2018;191:234–258. [ DOI ] [ PubMed ] [ Google Scholar ] 79. Davidoff  E, Reifenstein  EC, Jr. The results of eighteen months of Benzedrine sulfate therapy in psychiatry. Am J Psychiatry.  1939;95:945–970. [ Google Scholar ] 80. Kornetsky  C.  Hyporesponsivity of chronic schizophrenic patients to dextroamphetamine. Arch Gen Psychiatry.  1976;33:1425–1428. [ DOI ] [ PubMed ] [ Google Scholar ] 81. van Kammen  DP, Bunney  WE, Jr, Docherty  JP, et al.  d-Amphetamine-induced heterogeneous changes in psychotic behavior in schizophrenia. Am J Psychiatry.  1982;139:991–997. [ DOI ] [ PubMed ] [ Google Scholar ] 82. Kraemer  HC, Kupfer  DJ.  Size of treatment effects and their importance to clinical research and practice. Biol Psychiatry.  2006;59:990–996. [ DOI ] [ PubMed ] [ Google Scholar ] 83. Fisher  M, Nahum  M, Howard  E, et al.  Supplementing intensive targeted computerized cognitive training with social cognitive exercises for people with schizophrenia: An interim report. Psychiatr Rehabil J.  2017;40:21–32. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 84. Nuechterlein  KH, McEwen  SC, Ventura  J, et al.  Aerobic exercise enhances cognitive training effects in first-episode schizophrenia: randomized clinical trial demonstrates cognitive and functional gains. Psychol Med.  2023;53:4751–4761. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 85. Vinogradov  S, Fisher  M, Warm  H, Holland  C, Kirshner  MA, Pollock  BG.  The cognitive cost of anticholinergic burden: decreased response to cognitive training in schizophrenia. Am J Psychiatry.  2009;166:1055–1062. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 86. Joshi  YB, Thomas  ML, Braff  DL, et al.  Comprehensive characterization of anticholinergic medication burden-associated cognitive impairment in schizophrenia. Am J Psychiatry.  2021;178:838–847. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 87. Noetzli  M, Eap  CB.  Pharmacodynamic, Pharmacokinetic and Pharmacogenetic Aspects of Drugs Used in the Treatment of Alzheimer’s Disease. Clin Pharmacokinet.  2013;52:225–241. [ DOI ] [ PubMed ] [ Google Scholar ] 88. Hermes  EDA, Sokoloff  D, Stroup  TS, Rosenheck  RA.  Minimum clinically important difference in the Positive and Negative Syndrome Scale with data from the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE). J Clin Psychiatry.  2012;73:526–532. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 89. Hjorthøj  C, Stürup  AE, McGrath  JJ, Nordentoft  M.  Years of potential life lost and life expectancy in schizophrenia: a systematic review and meta-analysis. Lancet Psychiatry. 2017;4:295–301. [ DOI ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials sbaf015_suppl_Supplementary_Tables_S1-S9_Figures_S1-S9 sbaf015_suppl_supplementary_tables_s1-s9_figures_s1-s9.docx (335.6KB, docx) Data Availability Statement Raw T1 and P30 data for all PANSS measures among “completers” (n = 50) used to calculate within-subject effect sizes are found in Supplemental Results ( Table S5 ); remaining data is available on request, pending publication of findings. Articles from Schizophrenia Bulletin are provided here courtesy of Oxford University Press ACTIONS View on publisher site PDF (1.1 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top