Increasing stimulus similarity drives nonmonotonic representational change in hippocampus

Studies of hippocampal learning have obtained seemingly contradictory results, with manipulations that increase coactivation of memories sometimes leading to differentiation of these memories, but sometimes not. These results could potentially be reconciled using the nonmonotonic plasticity hypothesis, which posits that representational change (memories moving apart or together) is a U-shaped function of the coactivation of these memories during learning. Testing this hypothesis requires manipulating coactivation over a wide enough range to reveal the full U-shape. To accomplish this, we used a novel neural network image synthesis procedure to create pairs of stimuli that varied parametrically in their similarity in high-level visual regions that provide input to the hippocampus. Sequences of these pairs were shown to human participants during high-resolution fMRI. As predicted, learning changed the representations of paired images in the dentate gyrus as a U-shaped function of image similarity, with neural differentiation occurring only for moderately similar images.

Low-dimensional encoding of decisions in parietal cortex reflects long-term training history

Neurons in parietal cortex exhibit task-related activity during decision-making tasks. However, it remains unclear how long-term training to perform different tasks over months or even years shapes neural computations and representations. We examine lateral intraparietal area (LIP) responses during a visual motion delayed-match-to-category (DMC) task. We consider two pairs of monkeys with different training histories: one trained only on the DMC task, and another first trained to perform fine motion-direction discrimination. We introduce generalized multilinear models to quantify low-dimensional, task-relevant components in population activity. During the DMC task, we found stronger cosine-like motion-direction tuning in the pretrained monkeys than in the DMC-only monkeys, and that the pretrained monkeys' performance depended more heavily on sample-test stimulus similarity. These results suggest that sensory representations in LIP depend on the sequence of tasks that the animals have learned, underscoring the importance of training history in studies with complex behavioral tasks.

Increasing stimulus similarity drives nonmonotonic representational change in hippocampus

Studies of hippocampal learning have obtained seemingly contradictory results, with manipulations that increase coactivation of memories sometimes leading to differentiation of these memories, but sometimes not. These results could potentially be reconciled using the nonmonotonic plasticity hypothesis, which posits that representational change (memories moving apart or together) is a U-shaped function of the coactivation of these memories during learning. Testing this hypothesis requires manipulating coactivation over a wide enough range to reveal the full U-shape. To accomplish this, we used a novel neural network image synthesis procedure to create pairs of stimuli that varied parametrically in their similarity in high-level visual regions that provide input to the hippocampus. Sequences of these pairs were shown to human participants during high-resolution fMRI. As predicted, learning changed the representations of paired images in the dentate gyrus as a U-shaped function of image similarity, with neural differentiation occurring only for moderately similar images.

The Capacity and Organization of Gustatory Working Memory

Remembering a particular taste is crucial in food intake and associative learning. We investigated whether taste can be dynamically encoded, maintained, and retrieved on short time-scales consistent with working memory (WM). We used novel single and multi-item taste recognition tasks to investigate the organization and capacity of gustatory WM. In Experiment 1, we show that a single taste can be reliably recognized despite multiple oro-sensory interferences suggesting active and resilient maintenance. When multiple tastes were presented, the resolution with which these could be maintained, depended on their serial position implying a role of attention. Participants reliably recognized up to three tastes, compatible with a limited capacity of gustatory WM. Lastly, recognition was better for match than foil trials likely due to increased stimulus similarity in foil trials. Together, the results advocate a hybrid model of gustatory WM with a limited number of slots where items are stored with varying precision.

Q-SpAM: How to Efficiently Measure Similarity in Online Research

Measuring the similarity of stimuli is of great interest to a variety of social scientists. Spatial arrangement by dragging and dropping “more similar” targets closer together on the computer screen is a precise and efficient method to measure stimulus similarity. We present Qualtrics-spatial arrangement method (Q-SpAM), a feature-rich and user-friendly online version of spatial arrangement. Combined with crowdsourcing platforms, Q-SpAM provides fast and affordable access to similarity data even for large stimulus sets. Participants may spatially arrange up to 100 words or images, randomly selected targets, self-selected targets, self-generated targets, and targets self-marked in different colors. These and other Q-SpAM features can be combined. We exemplify how to collect, process, and visualize similarity data with Q-SpAM and provide R and Excel scripts to do so. We then illustrate Q-SpAM’s versatility for social science, concluding that Q-SpAM is a reliable and valid method to measure the similarity of lots of stimuli with little effort.

When “capacity” changes with set size: Ensemble representations support the detection of across-category changes in visual working memory

Is there a fixed limit on how many objects we can hold actively in mind? Generally, researchers have found participants are able to remember fewer objects if they are more complex, suggesting a limited resource rather than a fixed number of objects best explains working memory performance. However, some evidence has suggested that stimulus similarity better accounts for these effects, and that after accounting for such similarity, the data support a slot-based fixed- item limit for working memory. Much of the evidence used to support the latter claim relies on working memory displays containing different categories of items. It has been found that for large, across-category changes, performance does not differ for different kinds of complex stimuli. However, many of these studies fail to adequately control for the potential use of ensemble information in discriminating such large, across-category changes. Here, we sought to identify how much ensemble representations may explain performance across these tasks. In Experiment 1, we observed that as set size increased from 4 to 12 items, capacity estimates for across-category changes increased linearly as well, providing evidence against the claim of a fixed capacity. In Experiment 2, we controlled for stimulus complexity and similarity but varied the utility of ensemble representations for the change detection task. We observed significantly greater capacity when ensemble information could be used. Altogether, these results are contrary to a slot-like, fixed-object constraint on working memory capacity, and consistent with object complexity and ensemble representations strongly affecting working memory performance.

Experience-dependent effects of muscimol-induced hippocampal excitation on mnemonic discrimination

Memory requires similar episodes with overlapping features to be represented distinctly, a process that is disrupted in many clinical conditions as well as normal aging. Data from humans have linked this ability to activity in hippocampal CA3 and dentate gyrus (DG). While animal models have shown the perirhinal cortex is critical for disambiguating similar stimuli, hippocampal activity has not been causally linked to discrimination abilities. The goal of the current study was to determine how disrupting CA3/DG activity would impact performance on a rodent mnemonic discrimination task. Rats were surgically implanted with bilateral guide cannulae targeting dorsal CA3/DG. In Exp.1, the effect of intra-hippocampal muscimol on target-lure discrimination was assessed within subjects in randomized blocks. Muscimol initially impaired discrimination across all levels of target-lure similarity, but performance improved on subsequent test blocks irrespective of stimulus similarity and infusion condition. To clarify these results, Exp.2 examined whether prior experience with objects influenced the effect of muscimol on target-lure discrimination. Rats that received vehicle infusions in a first test block, followed by muscimol in a second block, did not show discrimination impairments for target-lure pairs of any similarity. In contrast, rats that received muscimol infusions in the first test block were impaired across all levels of target-lure similarity. Sustained effects of muscimol in disrupting behavioral performance after repeated infusions were verified in a spatial alternation task. At the conclusion of behavioral experiments, fluorescence in situ hybridization for the immediate-early genes Arc and Homer1a was used to determine the proportion of neurons active following muscimol infusion. Contrary to expectations, muscimol increased neural activity in DG. An additional experiment was carried out to quantify neural activity in naïve rats that received an intra-hippocampal infusion of vehicle or muscimol. Results confirmed that muscimol led to DG excitation, likely through its actions on interneuron populations in hilar and molecular layers of DG and consequent disinhibition of principal cells. Taken together, our results suggest disruption of coordinated neural activity across the hippocampus impairs mnemonic discrimination when lure stimuli are novel.

Exploring the Effect of Stimulus Similarity on the Summation Effect in Causal Learning

Several contemporary models anticipate that the summation effect is modulated by the similarity between the cues forming a compound. Here, we explore this hypothesis in a series of causal learning experiments. Participants were presented with two visual cues that separately predicted a common outcome and later asked for the outcome predicted by the compound of the two cues. Similarity was varied between groups through changes in shape, spatial position, color, configuration, and rotation. In variance with the predictions of these models, we observed similar and strong levels of summation in both groups across all manipulations of similarity. The effect, however, was significantly reduced by manipulations intended to impact assumptions about the causal independence of the cues forming the compound, but this reduction was independent of stimulus similarity. These results are problematic for similarity-based models and can be more readily explained by rational approaches to causal learning.