Biology subjects

Schmitt, L.-M.

Publications and source records attributed to Schmitt, L.-M..

9 recordsLinked to original sources

bioRxiv2026

Generative and discriminative recurrence employ opposing strategies for robust vision

Recurrence is thought to enhance the robustness of biological vision, but how it achieves this feat is largely unknown. Perceptual robustness can be implemented through either lateral connections supporting local integration within a processing stage or feedback connections drawing on broader context from higher stages, and through either a discriminative objective optimising task-relevant classification or a generative objective learning to reconstruct the causes of visual input. But do these different types of recurrence engage distinct computational strategies? As this question is difficult to test in vivo, we endowed convolutional neural networks with varying recurrent architectures and training objectives, and evaluated the consequences for internal representations and behaviour across noise levels. Two distinct computational strategies emerged. Generative feedback followed a reductionist strategy, with representations becoming lower-dimensional through denoising, achieving robustness at moderate noise levels without noise training. Both discriminative lateral and feedback recurrence followed an expansionist strategy, increasing dimensionality to sharpen discriminability without denoising, but requiring noise training to achieve robustness. These dissociable signatures reflect fundamentally different computational mechanisms of robust vision and provide testable predictions for which form of recurrence the brain employs.

neuroscience↗

bioRxiv2025

Narrative Context Shifts Gaze from Visual to Semantic Salience

Humans make over a hundred thousand eye movements daily to gather visual information. But what determines where we look? Current computational models typically link gaze behaviour to visual features of isolated images, but we know that eye movements are also strongly shaped by cognitive goals: Observers gather information that helps them to understand, rather than just represent, the world. Within this framework, observers should focus more on information that updates ones understanding of the environment, and less on what is purely visually salient. Here we tested this hypothesis using a free-viewing paradigm of narratives where we experimentally manipulated the meaningfulness of temporal context by either presenting pictures in a coherent, i.e. correct, order, or in a temporally shuffled order. We developed a novel approach to quantify which visual information is semantically salient (i.e., important for understanding): we separately obtained language narratives for images in stories, and computed the contextual surprisal of visual objects using a large language model. The ability of this semantic saliency model in explaining gaze behaviour was compared to a state-of-the art model of visual saliency (DeepGaze-II). We found that individuals looked relatively more often and more quickly at semantically salient objects when images were presented in coherent compared to shuffled order. In contrast, visual salience did not better account for gaze behaviour in coherent than shuffled order. These findings highlight how internal contextual models guide visual sampling and demonstrate that language models could offer a powerful tool for capturing gaze behavior in richer, meaningful settings.

neuroscience↗

bioRxiv2025

Prior Information Shapes Perceptual Evidence Accumulation Dynamics Differentially in Psychosis

Humans rely on prior information to navigate sensory uncertainty: Such priors could shape the decision process before sensory evidence is gathered (origin model), or could amplify sensory evidence dynamically (gain model). Dysfunctionalities in the utilisation of priors may underlie hallucinatory percepts and delusional ideation in psychosis, yet their impact on decision-making across sensory modalities has remained unclear. Using a perceptual target-detection task across auditory and visual domains in laboratory and online samples, we applied hierarchical drift diffusion modelling to examine how prior probabilities shape evidence accumulation in clinical and non-clinical populations. We show that in healthy individuals, prior information enhances sensory gain and decision flexibility, as represented by drift criterion and rate, consistent with the gain model. In contrast, individuals with psychosis exhibit diminished sensory gain, relying instead on pre-evidence biases, consistent with the origin model. Notably, greater positive symptom severity predicted a reduction in traditional criterion decision bias. These results suggest that sensory gain deficits may serve as a computational marker for psychosis progression, linking dysfunctional prior use to perceptual aberrancies. By demonstrating how prior information modulates evidence accumulation across sensory modalities, our study advances the understanding of psychotic perception and decision-making, offering insights for computational psychiatry and fine-tuned clinical diagnostics.

neuroscience↗

bioRxiv2024

Limited cognitive resources reduce the language predictability benefit across the adult lifespan

Humans routinely anticipate upcoming language, but whether such predictions come at a cognitive cost remains debated. In this study, we demonstrate the resource-dependent nature of predictive mechanisms in language comprehension across the lifespan: Experimentally limiting executive resources through a concurrent task reduces the effect of language predictability on reading time. Participants (N=175, replication N=96) read short articles presented word-by-word while completing a secondary font colour n-back task, thus varying cognitive demand. Language predictability was indexed by word surprisal as derived from a pre-trained large language model (GPT-2). Across two independent samples, our findings reveal that language predictions are not cost-free: They draw on executive control resources, and this dependency becomes more pronounced with age (18-85 years). These results help resolve the debate over cognitive demands in language comprehension and highlight prediction as a dynamic, resource-dependent process across the lifespan.

neuroscience↗

bioRxiv2021

Age-related differences in the neural network dynamics underlying the predictability gain

Speech comprehension is often challenged by increased background noise, but can be facilitated via the semantic context of a sentence. This predictability gain relies on an interplay of language-specific semantic and domain-general brain regions. However, age-related differences in the interactions within and between semantic and domain-general networks remain poorly understood. Using functional neuroimaging, we investigated commonalities and differences in network interactions enabling processing of degraded speech in healthy young and old participants. Participants performed a sentence repetition task while listening to sentences with high and low predictable endings and varying intelligibility. Stimulus intelligibility was adjusted to individual hearing abilities. Older adults showed an undiminished behavioural predictability gain. Likewise, both groups recruited a similar set of semantic and cingulo-opercular brain regions. However, we observed age-related differences in effective connectivity for high predictable speech of increasing intelligibility. Young adults exhibited stronger connectivity between regions of the cingulo-opercular network and between left insula and the posterior middle temporal gyrus. Moreover, these interactions were excitatory in young adults but inhibitory in old adults. Finally, the degree of the inhibitory influence between cingulo-opercular regions was predictive of the behavioural sensitivity towards changes in intelligibility for high predictable sentences in older adults only. Our results demonstrate that the predictability gain is relatively preserved in older adults when stimulus intelligibility is individually adjusted. While young and old participants recruit similar brain regions, differences manifest in underlying network interactions. Together, these results suggest that ageing affects the network configuration rather than regional activity during successful speech comprehension under challenging listening conditions.

neuroscience↗

bioRxiv2020

Predicting speech from a cortical hierarchy of event-based timescales

How can anticipatory neural processes structure the temporal unfolding of context in our natural environment? We here provide evidence for a neural coding scheme that sparsely updates contextual representations at the boundary of events and gives rise to a hierarchical, multi-layered organization of predictive language comprehension. Training artificial neural networks to predict the next word in a story at five stacked timescales and then using model-based functional MRI, we observe a sparse, event-based "surprisal hierarchy". The hierarchy evolved along a temporo-parietal pathway, with model-based surprisal at longest timescales represented in inferior parietal regions. Along this hierarchy, surprisal at any given timescale gated bottom-up and top-down connectivity to neighbouring timescales. In contrast, surprisal derived from a continuously updated context influenced temporo-parietal activity only at short timescales. Representing context in the form of increasingly coarse events constitutes a network architecture for making predictions that is both computationally efficient and semantically rich.

neuroscience↗

bioRxiv2020

Temporal selectivity declines in the aging human auditory cortex

Current models successfully describe the auditory cortical response to natural sounds with a set of spectra-temporal features. However, these models have hardly been linked to the ill-understood neurobiological changes that occur in the aging auditory cortex. Modelling the hemodynamic response to a rich natural sound mixture in N=64 listeners of varying age, we here show that in older listeners auditory cortex, the key feature of temporal rate is represented with a markedly broader tuning. This loss of temporal selectivity is most prominent in primary auditory cortex and planum temporale, with no such changes in adjacent auditory or other brain areas. Amongst older listeners, we observe a direct relationship between chronological age and temporal-rate tuning, unconfounded by auditory acuity or model goodness of fit. In line with senescent neural dedifferentiation more generally, our results highlight decreased selectivity to temporal information as a hallmark of the aging auditory cortex. Impact StatementAge-dependent broadening of tuning to temporal modulation rate in older humans cortical sound representations highlights decreased temporal selectivity as an anatomically and functionally specific hallmark of the aging auditory cortex.

neuroscience↗

bioRxiv2019

Neural modelling of the semantic predictability gain under challenging listening conditions

When speech intelligibility is reduced, listeners exploit constraints posed by semantic context to facilitate comprehension. The left angular gyrus (AG) has been argued to drive this semantic predictability gain. Taking a network perspective, we ask how the connectivity within language-specific and domain-general networks flexibly adapts to the predictability and intelligibility of speech. During continuous functional magnetic resonance imaging (fMRI), participants repeated sentences, which varied in semantic predictability of the final word and in acoustic intelligibility. At the neural level, highly predictable sentences led to stronger activation of left-hemispheric semantic regions including subregions of the AG (PGa, PGp) and posterior middle temporal gyrus when speech became more intelligible. The behavioural predictability gain of single participants mapped onto the same regions but was complemented by increased activity in frontal and medial regions. Effective connectivity from PGa to PGp increased for more intelligible sentences. In contrast, inhibitory influence from pre-supplementary motor area to left insula was strongest when predictability and intelligibility of sentences were either lowest or highest. This interactive effect was negatively correlated with behavioural predictability gain. Together, these results suggest that successful comprehension in noisy listening conditions relies on an interplay of semantic regions and concurrent inhibition of cognitive control regions when semantic cues are available.

neuroscience↗

bioRxiv2018

Does closing the eyes enhance auditory attention? Eye closure increases attentional alpha-power modulation but not listening performance

In challenging listening conditions, closing the eyes is a strategy with intuitive appeal to improve auditory attention and perception. On the neural level, closing the eyes increases the power of alpha oscillations ([~]10 Hz), which are a prime signature of auditory attention. Here, we test whether eye closure benefits neural and behavioural signatures of auditory attention and perception. Participants (N = 22) attended to one of two alternating streams of spoken numbers with open or closed eyes in a darkened chamber. After each trial, participants indicated whether probes had been among the to-be-attended or to-be-ignored numbers. In the electroencephalogram, states of relative high versus low alpha power accompanied the presentation of attended versus ignored numbers. Importantly, eye closure did not only increase the overall level of absolute alpha power but also the attentional modulation thereof. Behaviourally, however, neither perceptual sensitivity nor response criterion was affected by eye closure. To further examine whether this behavioural null-result would conceptually replicate in a simple auditory detection task, a follow-up experiment was conducted that required participants (N = 19) to detect a near-threshold target tone in noise. As in the main experiment, our results provide evidence for the absence of any difference in perceptual sensitivity and criterion for open versus closed eyes. In sum, we demonstrate here that the modulation of the human alpha rhythm by auditory attention is increased when participants close their eyes. However, our results speak against the widely held belief that eye closure per se improves listening behaviour.

neuroscience↗