Open Talks: Cognitive Theme
(Parallel Session 3 / Talk Session 2)
Speaker 1: Flexibility of emerging face categorization at different levels of abstraction
Miguel Granja Espírito Santo (1) and Johan Wagemans (1)
(1) KU Leuven
Categorization of visual stimuli at different levels of abstraction relies on the encoding of relevant diagnostic features present at different spatial scales. We used the Eidolon Factory, an image-manipulation algorithm that introduces random disarray fields across spatial scales, to study how such a process flexibly combines perceptual information to perform successful categorization depending on task demands. Images of animal faces, human faces and everyday objects were disarrayed coherently (random fields correlated) or incoherently (random fields randomized) to create a family of 50 eidolons per stimulus image with increasing disarray. Participants (N = 243) viewed each family of eidolons in a smooth sequence from maximum disarray to no disarray and performed a category verification task either at the superordinate (any face type) or basic (human face only) levels at two levels of uncertainty: Participants in one group used their gut feeling to respond, while another group had to be sure of their decision. When participants used their gut feeling to respond, we observed a superordinate-level advantage. When they were sure of their response, we observed a basic-level advantage. Coherently disarrayed sequences impaired target detection compared to incoherently disarrayed sequences for both levels of response certainty. Furthermore, participants’ sensitivity in the Any Face condition increased when they observed coherently disarrayed sequences and had to be sure of their response. These results suggest that the visual system does not strictly adhere to feedforward processing but flexibly adjusts to the relevant perceptual information depending on task context.
Speaker 2: Prediction learning in autism in a volatile environment: A behavioral and multi-modal neuroimaging approach
Laurie-Anne Sapey-Triomphe (1), Joke Temmerman (1), Lauren Pattyn (1) and Johan Wagemans (1)
(1) KU Leuven
The social and non-social symptoms encountered in Autism Spectrum Disorder (ASD) could arise from atypical perceptual learning, according to Bayesian theories. Using behavior, model-based fMRI and Magnetic Resonance Spectroscopy, we aimed to characterize prediction learning in a volatile context in ASD. Twenty-six neurotypical (NT) adults and 26 adults with ASD performed an associative learning task in the MR scanner, based on Weilnhammer et al. (2018). After hearing a low or high tone, they predicted the rotation direction of two dots and reported their percept. The tone and rotation were congruent or incongruent in 75% and 12.5% of the trials, respectively. The association reversed across time. In another 12.5% of the trials (ambiguous trials), the dot pair did not rotate but could be perceived as rotating if expectations biased percepts. Both groups managed to learn the tone-rotation association and tended to be biased by their expectations in ambiguous trials. In ASD only, the percentage of correct predictions negatively correlated with the GLX/GABA ratio in an inferior frontal region. High-level predictions involved a large neural network in both groups, including inferior and middle frontal regions, the hippocampus and the cingulate cortex. Low-level predictions only involved visual regions in NT and an auditory region in ASD. There were no significant group differences at the brain level. To conclude, adults with ASD can learn predictions in a volatile environment and rely on a relatively similar neural network, but may have different underlying molecular mechanisms.
Speaker 3: A diminished neural response to one’s own face in Autism, as shown using fast periodic visual stimulation
Annabel Nijhof (1,2), Caroline Catmur (2), Rebecca Brewer (3), Roeljan Wiersema (1), Geoff Bird (2,4)
(1) UGent; (2) King's College London, UK; (3) Royal Holloway, UK; (4) University of Oxford
Research suggests that self-related processing is altered in autism, with studies reporting a diminished neural response to their own name and face in individuals with autism. Although EEG data are generally quite noisy, with Fast Periodic Visual Stimulation (FPVS) one can quickly obtain a lot of data with a high signal-to-noise ratio. We used FPVS to test discrimination of one’s own face among other faces in adults with and without autism, to further investigate self-processing in autism. EEG was recorded for 20 adults with autism and 24 typically developed adults, while they were watching rapidly alternating naturalistic face images. Strangers’ faces were presented at a base frequency of 5.77 Hz, in three runs of 82 seconds. Every fifth image (oddball frequency: 1.154 Hz) had the same identity (own, close other or stranger). Signal-to-noise ratio at the oddball frequency was highest in two parieto-occipital and a mid-frontal cluster. Within each cluster, the control group showed a significantly greater response to one’s own than to the other faces, and greater for the close other’s than the stranger’s face. In contrast, in the autism group, familiar faces elicited stronger responses than the stranger’s face, but the difference between the own and close other’s face was not significant. In under five minutes, we extended previous findings of a diminished response to one’s own face in adults with autism, providing further evidence for altered self-processing in autism. More generally, FPVS can be considered promising for studying face recognition in clinical groups.
Speaker 4: Contentless thinking is associated with whole-brain positive inter-areal connectivity patterns
Sepehr Mortaheb (1), Laurens Van Calster (1), Paradeisios Alexandros Boulakis (1),
Kleio Georgoula (1), Steve Majerus (1) and Athena Demertzi (1)
Kleio Georgoula (1), Steve Majerus (1) and Athena Demertzi (1)
During spontaneous mentation, our minds are occupied with different contents, including periods of contentless thinking (mind blanking (MB)). As the frequency of MB events is non-negligible, an emerging question is whether this mental state constitutes an accidental blip or rather a default function of our ongoing mental flow. Using fMRI experience-sampling in 36 typical subjects during which MB could be chosen among various mental states, we show that MB is less frequent (5.75%) compared to sensory-oriented (Sens, M=19.79%), stimulus-dependent (SDep, 31.57%), and stimulus-independent thoughts (SInd, 42.90%), distributed equally across time (Chi-square uniformity test, x2=12.31, p=0.20). The probability of reporting mind blanking is low but equal when departing from the other states (Markov chain transition probability=0.06), suggesting that this state is not driven by specific mental content. FMRI phase-based coherence showed that a recurrent brain pattern of overall positive functional connectivity was closer to mind blanking reports (in the sense of cosine distance) compared to other mental states (p=0.03 for MB vs. Sens and p=0.003 for MB vs. SDep and p<0.001 for MB vs. SInd, generalized linear mixed effect model and posthoc Tukey's test). This indicates that mind blanking is a default mental state supported by an over-connected brain configuration. Such overall positive connectivity can reflect a distributed fight of multiple local units to enter into the supervisory attentional system, which may hinder reportable mental content formation.