S66. USING UNITIZATION TO CIRCUMVENT RELATIONAL MEMORY DEFICITS IN SCHIZOPHRENIA
Bibliographic record
Abstract
Abstract Background Relational memory (RM) is disproportionately affected in schizophrenia. Unitization is a cognitive strategy that compensates for RM deficits by facilitating the combination of disparate pieces of information to form a single functional unit. Unitization has been effective in circumventing RM impairment in hippocampal amnesia and older adults with RM deficits when performing a Transverse Patterning (TP) task. We developed a brief intervention using unitization to circumvent RM impairment in individuals with schizophrenia when learning arbitrary item-relations in a TP task. We subsequently developed the Relational Trip Task (RTT), which uses pairs of real-life stimuli (faces, places, objects) and a narrative frame for encoding to investigate if unitization generalizes to a more relatable real-life context. Methods Twenty-two individuals with an RM deficit and a diagnosis of schizophrenia or related disorder were pseudo-randomized to either the unitization or control condition, from which 19 completed all TP tasks and 17 completed RTT. TP performance was measured at screening. TP and RTT task performances were measured pre-post learning unitization. During the RTT-unitized task, participants created their own unitizations, with assistance (50% of the task) and on their own. The control group received unitization training following study participation. The TP-unitized control group results were included in the analysis. Results TP task performance (percentage of correct trials) did not significantly differ between control (M = 49%, SD = 13%) and unitization groups (M= 60%, SD = 18%) at screening (t(17) = -1.506, p = 0.15). A 2-way mixed analysis of variance (ANOVA) did not reveal a group-task interaction for unitization and control group accuracy in the four TP versions (F(3,51) = 2.38, p = 0.08). A main effect of task (F(3, 51) = 9.43, p > 0.001) was decomposed using Tukey HSD pairwise post-hoc analyses and showed significantly higher TP task accuracy following unitization (M = 85%, SD = 19%) compared to the TP task at screening (M = 55%, SD = 16% p > 0.001), before unitization (M = 61%, SD = 19%, p = 0.001) and when prompting self-unitization (M = 68%, SD = 28%, p = 0.04). No other significant differences in task accuracies were revealed. Group accuracies in the RTT were compared using a 2-way mixed ANOVA, and yielded a significant interaction between group and task accuracy (F(1,15) = 4.93, p = 0.042). Simple main effect analysis showed that accuracy in the RTT post unitization training (M = 90%, SD = 9%) was higher than before training (M = 77%, SD = 14%, p = 0.046), but performance between the same versions of the RTT did not significantly differ in the control group (M1 = 73%, SD = 19%; M2 = 70%, SD = 23%; p = 0.26). Discussion TP performance improved when the unitization strategy was provided, but not when the self-generation of unitization was encouraged. Improved RTT performance was limited to the unitization group, suggesting that effects were unitization-specific rather than lead by practice. Logic follows that this strategy may be generalizable to more relatable, real-life contexts. Self-generation of unitization was effective in improving task performance when assistance was provided rather than merely encouraged, suggesting that patients may benefit from guidance generating their own unitizations rather than integrating this strategy on their own. These findings should be replicated in a larger sample, and strategies to provide effective self-generation of unitization should be explored. Moreover, the extent to which the RTT can detect differential relational memory impairment in individuals with schizophrenia when compared to healthy controls warrants further investigation.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.004 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".