Caveats and Nuances of Model-Based and Model-Free Representational Connectivity Analysis

Recurrent connectivity supports higher-level visual and semantic object representations in the brain

Visual object recognition has been traditionally conceptualised as a predominantly feedforward process through the ventral visual pathway. While feedforward artificial neural networks (ANNs) can achieve human-level classification on some image-labelling tasks, it's unclear whether computational models of vision alone can accurately capture the evolving spatiotemporal neural dynamics. Here, we probe these dynamics using a combination of representational similarity and connectivity analyses of fMRI and MEG data recorded during the recognition of familiar, unambiguous objects. Modelling the visual and semantic properties of our stimuli using an artificial neural network as well as a semantic feature model, we find that unique aspects of the neural architecture and connectivity dynamics relate to visual and semantic object properties. Critically, we show that recurrent processing between the anterior and posterior ventral temporal cortex relates to higher-level visual properties prior to semantic object properties, in addition to semantic-related feedback from the frontal lobe to the ventral temporal lobe between 250 and 500 ms after stimulus onset. These results demonstrate the distinct contributions made by semantic object properties in explaining neural activity and connectivity, highlighting it as a core part of object recognition not fully accounted for by current biologically inspired neural networks.

Altered amygdalar emotion space in borderline personality disorder normalizes following dialectical behaviour therapy

BACKGROUND

Borderline personality disorder (BPD) is a mental health condition characterized by an inability to regulate emotions or accurately process the emotional states of others. Previous neuroimaging studies using classical univariate analyses have tied such emotion dysregulation to aberrant activity levels in the amygdala of patients with BPD. However, multivariate analyses have not yet been used to investigate how representational spaces of emotion information may be systematically altered in patients with BPD.

METHODS

Patients with BPD performed an emotional face matching task while undergoing MRI before and after a 10-week inpatient program of dialectical behavioural therapy. Representational similarity analysis (RSA) was applied to activity patterns (evoked by angry, fearful, neutral and surprised faces) in the amygdala and temporo-occipital fusiform gyrus of patients with BPD and in the amygdala of healthy controls.

RESULTS

We recruited 15 patients with BPD (8 females, 6 males, 1 transgender male) to participate in the study, and we obtained a neuroimaging data set for 25 healthy controls for a comparative analysis. The RSA of the amygdala revealed a negative bias in the underlying affective space (in that activity patterns evoked by angry, fearful and neutral faces were more similar to each other than to patterns evoked by surprised faces), which normalized after therapy. This bias-to-normalization effect was present neither in activity patterns of the temporo-occipital fusiform gyrus of patients nor in amygdalar activity patterns of healthy controls.

LIMITATIONS

Larger samples and additional questionnaires would help to better characterize the association between specific aspects of therapy and changes in the neural representational space.

CONCLUSION

Our findings suggest a more refined role for the amygdala in the pathological processing of perceived emotions and may provide new diagnostic and prognostic imaging-based markers of emotion dysregulation and personality disorders.Clinical trial registration: DRKS00019821, German Clinical Trials Register (Deutsches Register Klinischer Studien).