Biology subjects

Rissman, J.

Publications and source records attributed to Rissman, J..

6 recordsLinked to original sources

bioRxiv2026

Spectral and non-spectral EEG measures in the prediction of working memory task performance and psychopathology

Working memory (WM) supports the temporary maintenance of goal-relevant information and is disrupted across many neuropsychiatric disorders. We examined whether scalp electroencephalography (EEG) data features beyond spectral power, including waveform shape, broadband spectral structure, and signal complexity, provide complementary information for predicting cognitive and clinical outcomes. EEG was recorded from 200 adults spanning a broad range of neuropsychiatric symptom severity while they completed three WM task paradigms: Sternberg spatial WM (SWM), delayed face recognition (DFR), and dot pattern expectancy (DPX). Separate machine learning models were trained on EEG features from the encoding, delay, and probe phase of each task to predict participants task accuracy, reaction time (RT) variability, WM capacity, and psychopathology scores (Brief Psychiatric Rating Scale). A split-half analytic framework was used, with cross-validated model development in an exploratory dataset (N=100) and evaluation of statistically significant models in a held-out validation dataset (N=100). In the exploratory dataset, SWM task data best predicted WM capacity, DPX task data predicted RT variability, and DFR task data predicted psychopathology, suggesting that these three WM paradigms engage distinct neural processes relevant to different outcomes. No models reliably predicted task accuracy. Models incorporating features beyond spectral power generally outperformed power-only models, and task-derived features outperformed resting-state-derived features. However, only those models predicting WM capacity and RT variability generalized to the validation dataset; models predicting psychopathology did not. These findings demonstrate functional heterogeneity across WM paradigms, show that complementary EEG features enhance predictive modeling, and highlight the importance of rigorous validation for identifying robust brain-behavior relationships.

neuroscience↗

bioRxiv2026

Spontaneous emergence of context-dependent statistical learning in humans and neural networks

Humans readily extract statistical regularities from experience, yet natural environments require flexible adaptation when associative structures shift across changing contexts, often without warning. Across two experiments, we show that humans can incidentally learn overlapping and conflicting visual associations even when contexts dynamically alternate and remain unsignaled or only minimally cued. To probe the computational mechanisms supporting this adaptive capacity, we trained recurrent neural networks with gated recurrent units on the same statistical learning task without providing any explicit context information. These models spontaneously developed distributed internal representations that robustly separated conflicting associations and supported rapid adaptation to latent context shifts. Critically, we show that these distributed representations, strongly shaped by the models initial weight configuration, played a key role in preventing catastrophic interference between contexts. Together, these behavioral and computational results significantly advance our understanding of how humans and artificial systems can successfully learn and flexibly retrieve context-dependent associations under challenging conditions.

neuroscience↗

bioRxiv2025

Behavioural separation of face memory and face perception

A long-standing debate in neuropsychology concerns whether perception and memory function as independent systems or interact to support cognition. To investigate this, we developed the Face Memory and Perception (FMP) task, a novel paradigm designed to systematically disentangle whether and how these processes interact under different conditions. Across five independent datasets with over 900 participants in total, we observed consistent evidence that face perception and working memory operate independently when task demands are low, but in more complex conditions, these processes appear to interact. Notably, this interaction emerged only when the interference directly involved face-processing mechanisms, and did not arise from a general increase in cognitive load. Rather than the use of shared resources by overlapping cognitive processes, this interaction was driven by a shift in behavioural strategy from holistic to feature-based face processing as a result of maintenance-disrupting interference. These results underscore the fundamental independence of perception and working memory while also explaining some of the conditions under which interactions might be observed.

neuroscience↗

bioRxiv2024

Overcoming Limitations to Deep Learning in Domesticated Animals with TrioTrain

Variant calling across diverse species remains challenging as most bioinformatics tools default to assumptions based on human genomes. DeepVariant (DV) excels without joint genotyping while offering fewer implementation barriers. However, the growing appeal of a "universal" algorithm has magnified the unknown impacts when used with non-human genomes. Here, we use bovine genomes to assess the limits of human-genome-trained models in other species. We introduce the first multi-species DV model that achieves a lower Mendelian Inheritance Error (MIE) rate during single-sample genotyping. Our novel approach, TrioTrain, automates extending DV for species without Genome In A Bottle (GIAB) resources and uses region shuffling to mitigate barriers for SLURM-based clusters. To offset imperfect truth labels for animal genomes, we remove Mendelian discordant variants before training, where models are tuned to genotype the offspring correctly. With TrioTrain, we use cattle, yak, and bison trios to build 30 model iterations across five phases. We observe remarkable performance across phases when testing the GIAB human trios with a mean SNP F1 score >0.990. In HG002, our phase 4 bovine model identifies more variants at a lower MIE rate than DeepTrio. In bovine F1-hybrid genomes, our model substantially reduces inheritance errors with a mean MIE rate of 0.03 percent. Although constrained by imperfect labels, we find that multi-species, trio-based training produces a robust variant calling model. Our research demonstrates that exclusively training with human genomes restricts the application of deep-learning approaches for comparative genomics.

genomics↗

bioRxiv2023

Putting cognitive tasks on trial: A measure of reliability convergence

The surge in interest in individual differences has coincided with the latest replication crisis centered around brain-wide association studies of brain-behavior correlations. Yet the reliability of the measures we use in cognitive neuroscience, a crucial component of this brain-behavior relationship, is often assumed but not directly tested. Here, we evaluate the reliability of different cognitive tasks on a large dataset of over 250 participants, who each completed a multi-day task battery. We show how reliability improves as a function of number of trials, and describe the convergence of the reliability curves for the different tasks, allowing us to score tasks according to their suitability for studies of individual differences. To improve the accessibility of these findings, we designed a simple web-based tool that implements this function to calculate the convergence factor and predict the expected reliability for any given number of trials and participants, even based on limited pilot data.

neuroscience↗

bioRxiv2023

Establishing a causal role for left ventrolateral prefrontal cortex in value-directed memory encoding with high-definition transcranial direct current stimulation

One critical approach for promoting the efficiency of memory is to adopt selective encoding strategies to prioritize more valuable information. Past neuroimaging studies have shown that value-directed modulation of verbal memory depends heavily on engagement of left-lateralized semantic processing regions, particularly in ventrolateral prefrontal cortex (VLPFC). In the present study, we used high-definition direct current stimulation (HD-tDCS) to seek evidence for a causal role of left VLPFC in supporting the memory advantage for high-value items. Three groups of healthy young adult participants were presented with lists of words to remember, with each word accompanied by an arbitrarily assigned point value. During the first session, all participants received sham stimulation as they encoded five lists of 30 words each. Two of these lists were immediately tested with free recall, with feedback given to allow participants to develop metacognitive insight and strategies to maximize their point total. The second session had the exact same structure as the first, but the groups differed in whether they received continued sham stimulation (N=22) or anodal stimulation of the left VLPFC (N=21) or right VLPFC (N=20). Those lists not tested with immediate recall were tested with recognition judgments after a one-day delay. Since no brain stimulation was applied during this Day 2 test, any performance differences can be attributed to the effects of stimulation on Day 1 encoding processes. Anodal stimulation of left VLPFC significantly boosted participants memory encoding selectivity. In comparison, no such effect was seen in participants who received right VLPFC or sham stimulation. Estimates of recollection- and familiarity-based responding revealed that left VLPFC stimulation specifically amplified the effects of item value on recollection. These results demonstrate a causal role for left VLPFC in the implementation of selective value-directed encoding strategies, putatively by boosting deep semantic processing of high-value words. Our findings also provide further evidence on the hemispheric lateralization of value-directed verbal memory encoding.

neuroscience↗