Meijia Li

Background: Apathy, depression and anhedonia are clinically overlapping constructs, which hinders diagnostic clarity and treatment development. This study aimed to comprehensively characterise these syndromes to identify a core set of non-redundant symptoms that maximally dissociate them and to investigate the psychological nature of key distinguishing features. Methods: Data from seven datasets (N=4578) of healthy individuals and patients with major depressive disorder were analysed using the Apathy Motivation Index, Beck Depression Inventory and Snaith-Hamilton Pleasure Scale. A machine-learning algorithm identified the most informative, non-redundant items for dissociating ‘pure’ apathy, depression and anhedonia. The nature of emotional apathy was further investigated with follow-up studies. Results: Although substantial symptom overlap existed, ‘pure’ syndromes were present. Factor analysis revealed a robust five-factor structure, separating depression, anhedonia and three distinct apathy domains (behavioural, social and emotional). Machine learning identified 10 core symptoms that differentiated the pure syndromes with high accuracy (area under the curve >0.90) and could also identify well the presence of each syndrome in individuals suffering from two or more syndromes. Emotional apathy negatively correlated with depression and was specifically associated with reduced affective empathy and a diminished sensitivity to the intensity of negative facial emotions, rather than with alexithymia or antidepressant-induced emotional blunting. Conclusions: Apathy, depression and anhedonia are dissociable constructs with distinct symptom signatures. Emotional apathy is a unique dimension which provides a novel target for research. A 10-item Apathy-Depression-Anhedonia Measure developed here provides a pragmatic tool for rapid, precise phenotyping to guide more personalised therapeutic strategies.

Background Apathy, depression, and anhedonia are clinically overlapping constructs, which hinders diagnostic clarity and treatment development. This study aimed to comprehensively characterise these syndromes to identify a core set of non-redundant symptoms that maximally dissociate them and to investigate the psychological nature of key distinguishing features.Methods Data from seven datasets (N=4,578) of healthy individuals and patients with major depressive disorder were analysed using the Apathy Motivation Index, Beck Depression Inventory, and Snaith-Hamilton Pleasure Scale. A machine-learning algorithm identified the most informative, non-redundant items for dissociating "pure" apathy, depression, and anhedonia. The nature of emotional apathy was further investigated with follow-up studies.Results Although substantial symptom overlap existed, "pure" syndromes were present. Factor analysis revealed a robust five-factor structure, separating depression, anhedonia, and three distinct apathy domains (behavioural, social, emotional). Machine learning identified 10 core symptoms that differentiated the pure syndromes with high accuracy (AUC >0.90) and could also identify well the presence of each syndrome in individuals suffering from two or more syndromes. Emotional apathy negatively correlated with depression and was specifically associated with reduced affective empathy and a diminished sensitivity to the intensity of negative facial emotions, rather than with alexithymia or antidepressant-induced emotional blunting.Conclusions Apathy, depression, and anhedonia are dissociable constructs with distinct symptom signatures. Emotional apathy is a unique dimension which provides a novel target for research. A 10-item Apathy-Depression-Anhedonia Measure (ADAM) developed here provides a pragmatic tool for rapid, precise phenotyping to guide more personalised therapeutic strategies.

Schizophrenia (ScZ) is characterized by prominent perceptual abnormalities. A deeper understanding of the neural mechanisms underlying these abnormalities is crucial for developing precise treatment strategies. Our study aimed to address the following primary questions. First, the functional role of various sub-oscillations within the alpha band remains unclear. Second, we aimed to identify biomarkers for the diagnostic purposes of ScZ. Third, the broader question of whether the diagnostic biomarker can also function as a treatment biomarker remains unknown. Resting-state EEG data from 55 ScZ patients and 61 healthy controls were analyzed to compare different sub-oscillations in the alpha band and their correlation with clinical symptoms (as measured by the general psychopathology scale). We discovered that distinct topographic patterns in low (~8 Hz) and high (~12 Hz) alpha may serve specific diagnostic and evaluative purposes respectively. Moreover, a pronounced gender bias was also observed. Low-alpha-band activity appeared to have more diagnostic relevance in females. On the other hand, the high-alpha difference was more relevant for evaluating the severity of symptoms in ScZ males. Our research has brought new insights into the neural oscillation mechanism of schizophrenia, which could substantially assist the formulating diagnosis of ScZ and the development of its treatment strategies.

Social navigation involves observing individuals navigating a socially rich environment sequentially while mentalizing their goals or preferences. The posterior cerebellum plays a central role in regulating human social behaviour and guiding social navigation. However, the extent to which cerebello-cerebral pathways mediate this role has not been sufficiently studied. Using Dynamic Causal Modelling on four novel social navigation studies (N = 107), our results showed many bidirectional connections between cerebellar Crus 2 and cortical mentalizing areas, including the temporoparietal junction, precuneus, and medial prefrontal cortex (mPFC), spanning both ipsilaterally and contralaterally. Connectivity patterns were only slightly modulated by experimental manipulations involving (i) active encoding versus passive observation of navigational sequences and (ii) social versus non-social sequences. Furthermore, when participants observed behaviours defying social expectations or norms, upward connectivity from the right Crus 2 to the ventral mPFC and downward connectivity from the precuneus to the left Crus 2 became stronger, while upward connectivity from the left Crus 2 to the ventral mPFC weakened. This study enhances our understanding of the role of the cerebellum in effective connectivity and its changes in response to social violations during social navigation.

The posterior cerebellum and anterior basal ganglia are critical subcortical structures for learning and identifying dynamic action sequences, in concert with the neocortex. The present analysis investigates the role of action sequences during social mentalizing, termed here dynamic or sequential social mentalizing. Although the role of the cerebellum in dynamic social mentalizing was extensively investigated during the last decade, the basal ganglia were long ignored. We conducted an activation likelihood estimation coordinate‐based meta‐analysis of sequential social mentalizing tasks (with 485 participants in 17 studies). These tasks required participants to make social mentalizing inferences ranging from low‐level goals to high‐level beliefs and traits, while either memorizing, generating or predicting temporal sequences of the social actions involved (i.e., social sequencing condition), or not (i.e., social non‐sequencing control condition), or did so for nonsocial objects (i.e., nonsocial sequencing control condition). The tasks also occasionally included inconsistencies in social behavior. Results revealed that the cerebellum exhibited a preference for social, sequencing, and inconsistent information, while the basal ganglia showed a preference for sequencing and inconsistency, without a general preference for social input. Meta‐analytic connectivity analysis further showed evidence of coactivation between mentalizing areas of the cerebellum, basal ganglia and cerebral neocortex. The present work underscores the role of subcortical structures in social mentalizing about dynamic action sequences.

Spatial trajectory planning and execution in a social context play a vital role in our daily lives. To study this process, participants completed a goal-directed task involving either observing a sequence of preferred goals and self-planning a trajectory (Self Sequencing) or observing and reproducing the entire trajectory taken by others (Other Sequencing). The results indicated that in the observation phase, witnessing entire trajectories created by others (Other Sequencing) recruited cerebellar mentalizing areas (Crus 2 and 1) and cortical mentalizing areas in the precuneus, ventral and dorsal medial prefrontal cortex and temporo-parietal junction more than merely observing several goals (Self Sequencing). In the production phase, generating a trajectory by oneself (Self Sequencing) activated Crus 1 more than merely reproducing the observed trajectories from others (Other Sequencing). Additionally, self-guided observation and planning (Self Sequencing) activated the cerebellar lobules IV and VIII more than Other Sequencing. Control conditions involving non-social objects and non-sequential conditions where the trajectory did not have to be (re)produced revealed no differences with the main Self and Other Sequencing conditions, suggesting limited social and sequential specificity. These findings provide insights into the neural mechanisms underlying trajectory observation and production by the self or others during social navigation.

Social norms are pivotal in guiding social interactions. The current study investigated the potential contribution of the posterior cerebellum, a critical region involved in perceiving and comprehending the sequential dynamics of social actions, in detecting actions that either conform to or deviate from social norms. Participants engaged in a goal-directed task in which they observed others navigating towards a goal. The trajectories demonstrated either norm-violating (trespassing forbidden zones) or norm-following behaviors (avoiding forbidden zones). Results revealed that observing social norm-violating behaviors engaged the bilateral posterior cerebellar Crus 2 and the right temporoparietal junction (TPJ) from the mentalizing network, and the parahippocampal gyrus (PHG) to a greater extent than observing norm-following behaviors. These mentalizing regions were also activated when comparing social sequences against non-social and non-sequential control conditions. Reproducing norm-violating social trajectories observed earlier, activated the left cerebellar Crus 2 and the right PHG compared to reproducing norm-following trajectories. These findings illuminate the neural mechanisms in the cerebellum associated with detecting norm transgressions during social navigation, emphasizing the role of the posterior cerebellum in detecting and signaling deviations from anticipated sequences.

Although the human cerebellum has a surface that is about 80% of that of the cerebral cortex and has about four times as many neurons, its functional organization is still very much uncharted. Despite recent attempts to provide resting-state and task-based parcellations of the cerebellum, these two approaches lead to large discrepancies. This article describes a comprehensive task-based functional parcellation of the human cerebellum based on a large-scale functional database, NeuroSynth, involving an unprecedented diversity of tasks, which were reliably associated with ontological key terms referring to psychological functions. Involving over 44,500 participants from this database, we present a parcellation that exhibits replicability with earlier resting-state parcellations across cerebellar and neocortical structures. The functional parcellation of the cerebellum confirms the major networks revealed in prior work, including sensorimotor, directed (dorsal) attention, divided (ventral) attention, executive control, mentalizing (default mode) networks, tiny patches of a limbic network, and also a unilateral language network (but not the visual network), and the association of these networks with underlying ontological key terms confirms their major functionality. The networks are revealed at locations that are roughly similar to prior resting-state cerebellar parcellations, although they are less symmetric and more fragmented across the two hemispheres. This functional parcellation of the human cerebellum and associated key terms can provide a useful guide in designing studies to test specific functional hypotheses and provide a reference for interpreting the results.