O4.3. PEOPLE WITH SCHIZOPHRENIA SHOW GREATER COGNITIVE COSTS OF STORING A SINGLE ITEM IN WORKING MEMORY

• Published in: Schizophrenia bulletin Vol. 45; no. Supplement_2; pp. S169 - S170
• Main Authors: Gold, James, Bansal, Sonia, Hahn, Britta, Robinson, Benjamin, Luck, Steven
• Format: Journal Article
• Language: English
• Published: US Oxford University Press 09.04.2019
• Subjects: Oral Abstracts; Schizophrenia
• ISSN: 0586-7614; 1745-1701
• DOI: 10.1093/schbul/sbz021.203

Abstract Background Working memory (WM) deficits are considered to be a core deficit in schizophrenia, implicated in the broad cognitive impairment seen in the illness. To date, the evidence supporting this view is based on correlations between measures of WM capacity or precision with other cognitive measures. Here we take an experimental approach and directly examine the impact of successful WM maintenance of a single, familiar item on the operation of other cognitive systems. Methods We studied 37 healthy controls (HCS) and 43 clinically stable people with schizophrenia (PSZ). Each trial consisted of a sequence of two potential target stimuli, T1 and T2. T1 was an X or a Y, presented for 100 ms. After delays varying from 100–800 ms, T2 was presented. T2 was a 1 or a 2 and required an immediate speeded discrimination response (one button for 1, another for 2). In one block of trials, subjects were instructed to ignore T1 but respond to T2 (attend-T2 trials). In another block, they were required to report T2 at the end of the trial, after making their speeded response to T1 (the attend-T1T2 condition). Thus, the attend-T1T2 condition required subjects to make a speeded response to T2 while holding T1 in WM. 240 trials were tested for each condition. We used median reaction times (RT) as our key dependent measure. The RT difference between attend-T1T2 and attend-T2 trials represents the cost of holding an item in WM. Results The main finding was that PSZ were dramatically slowed at responding to T2 when T1 was being held in WM (the attend-T1T2 condition). A repeated measures anova yielded main effects of group (PSZ slower), lag (RT longer at shortest T1-T2 intervals, and condition (attend-T1T2 slower than attend-T2; all ps<.001). There was also a significant group by condition interaction, F = 41.6, p<.001. Across lags, the mean difference score between attend-T1T2 RT and attend-T2 RT was nearly 3 times higher in PSZ than HCs (633 vs 219 msec). Performance in the T1T2 condition, but not the T2 condition correlated with measures of WM and general cognitive ability, with significantly larger correlations with WM storage capacity in PSZ than controls. Discussion These results open a new window into the consequences of WM impairment in PSZ. The process of maintaining a single, familiar item in WM substantially impacts choice reaction time. Given that many complex cognitive operations require the maintenance at least one item in WM while responding to other information, this laboratory finding may reveal a key factor underlying impaired cognitive performance in PSZ. The impairment documented here likely impacts the performance on widely used WM measures such as N-back tasks as well as complex span tasks. The results are consistent with other findings suggesting that PSZ may hyperfocus attentional and WM processes. Thus, PSZ overinvest resources in processing the T1 stimulus, thereby delaying the ability to respond to the T2 stimulus.