Automating the analysis of facial emotion expression dynamics: A computational framework and application in psychotic disorders.

Facial emotion expressions play a central role in interpersonal interactions; these displays are used to predict and influence the behavior of others. Despite their importance, quantifying and analyzing the dynamics of brief facial emotion expressions remains an understudied methodological challenge. Here, we present a method that leverages machine learning and network modeling to assess the dynamics of facial expressions. Using video recordings of clinical interviews, we demonstrate the utility of this approach in a sample of 96 people diagnosed with psychotic disorders and 116 never-psychotic adults. Participants diagnosed with schizophrenia tended to move from neutral expressions to uncommon expressions (e.g., fear, surprise), whereas participants diagnosed with other psychoses (e.g., mood disorders with psychosis) moved toward expressions of sadness. This method has broad applications to the study of normal and altered expressions of emotion and can be integrated with telemedicine to improve psychiatric assessment and treatment.

Striatal dopamine supports reward expectation and learning: A simultaneous PET/fMRI study.

Converging evidence from both human neuroimaging and animal studies has supported a model of mesolimbic processing underlying reward learning behaviors, based on the computation of reward prediction errors. However, competing evidence supports human dopamine signaling in the basal ganglia as also contributing to the generation of higher order learning heuristics. Here, we present data from a large (N = 81, 18-30yo), multi-modal neuroimaging study using simultaneously acquired task fMRI, affording temporal resolution of reward system function, and PET imaging with [11C]Raclopride (RAC), assessing striatal dopamine (DA) D2/3 receptor binding, during performance of a probabilistic reward learning task. Both fMRI activation and PET DA measures showed ventral striatum involvement for signaling rewards. However, greater DA release was uniquely associated with learning strategies (i.e., learning rates) that were more task-optimal within the best fitting reinforcement learning model. This DA response was associated with BOLD activation of a network of regions including anterior cingulate cortex, medial prefrontal cortex, thalamus and posterior parietal cortex, primarily during expectation, rather than prediction error, task epochs. Together, these data provide novel, human in vivo evidence that striatal dopaminergic signaling interacts with a network of cortical regions to generate task-optimal learning strategies, rather than representing reward outcomes in isolation.

Dissociation of basolateral and central amygdala effective connectivity predicts the stability of emotion-related impulsivity in adolescents and emerging adults with borderline personality symptoms: a resting-state fMRI study.

BACKGROUND: Borderline personality disorder (BPD) is associated with altered activity in the prefrontal cortex (PFC) and amygdala, yet no studies have examined fronto-limbic circuitry in borderline adolescents and emerging adults. Here, we examined the contribution of fronto-limbic effective connectivity (EC) to the longitudinal stability of emotion-related impulsivity, a key feature of BPD, in symptomatic adolescents and young adults. METHODS: We compared resting-state EC in 82 adolescents and emerging adults with and without clinically significant borderline symptoms (n BPD = 40, ages 13-30). Group-specific directed networks were estimated amongst fronto-limbic nodes including PFC, ventral striatum (VS), central amygdala (CeN), and basolateral amygdala (BLA). We examined the association of directed centrality metrics with initial levels and rates of change in emotion-related impulsivity symptoms over a one-year follow-up using latent growth curve models (LGCMs). RESULTS: In controls, ventromedial prefrontal cortex (vmPFC) and dorsal ACC had a directed influence on CeN and VS, respectively. In the BPD group, bilateral BLA had a directed influence on CeN, whereas in the healthy group CeN influenced BLA. LGCMs indicated that emotion-related impulsivity was stable across a one-year follow-up in the BPD group. Further, higher EC of R CeN to other regions in controls was associated with stronger within-person decreases in emotion-related impulsivity. CONCLUSIONS: Functional inputs from BLA and vmPFC appear to play competing roles in influencing CeN activity. In borderline adolescents and young adults, BLA may predominate over CeN activity, while in controls the ability of CeN to influence BLA activity predicted more rapid reductions in emotion-related impulsivity.

Borderline personality disorder: stress reactivity or stress generation? A prospective dimensional study.

BACKGROUND: Individuals diagnosed with borderline personality disorder (BPD) often describe their lives as stressful and unpredictable. However, it is unclear whether the adversity faced by those with BPD is a product of stress reactivity or stress generation. Here, we examined the dynamic, prospective associations between BPD and stressful life events over 3 years. Given the heterogeneity present in BPD, we sought to understand which empirically derived dimensions of this heterogeneous disorder explain stress reactivity v. stress generation. METHODS: Participants included 355 individuals diagnosed with BPD and followed longitudinally at three annual assessments. Auto-regressive cross-lagged panel models were used to examine prospective associations between stressful life events and three latent dimensions implicated in BPD: negative affect, disinhibition, and antagonism. RESULTS: Antagonism and disinhibition, but not negative affect, prospectively predicted dependent stressful life events (events the individual may have some role in). Evidence for decompensation under stress was more tenuous, with independent stressful life events (those presumably outside the individual's control) predicting increases in negative affect. CONCLUSIONS: Our longitudinal study of a well-characterized clinical sample found more evidence for stress generation than for stress-induced decompensation in BPD. Stress generation in BPD is driven by externalizing dimensions: antagonism and disinhibition. These results highlight the utility of empirically derived dimensions for parsing heterogeneity present in BPD, leading to improvements in diagnostic evaluation, clinical prediction, and individualized approaches to treatment planning.

Momentary interpersonal processes of suicidal surges in borderline personality disorder.

BACKGROUND: Suicide rates are high in borderline personality disorder (BPD) where interpersonal problems trigger intense affective dysregulation and impulses to act on suicidal thoughts. To date, however, no study has examined how interpersonal stressors contribute to momentary within-person links among affect and impulsivity with suicidal ideation (SI), and how those links vary over time in people's daily lives. METHODS: A total of 153 individuals diagnosed with BPD and 52 healthy controls completed a 21-day ecological momentary assessment protocol. Of these 153 individuals with BPD, 105 had a history of suicide attempts. Multilevel structural equation modeling was used to examine dynamic links among interpersonal perceptions, affect, state impulsivity, and suicidal intent. RESULTS: Aggregated across interactions, lower perceived warmth in others was associated with SI. This direct relationship, however, did not extend to momentary within-person associations. Instead, interpersonal conflicts were linked to SI indirectly via greater negative affect and lower positive affect. While a robust within-person link between interpersonal perceptions and impulsivity emerged, impulsivity did not account for the relationship between interpersonal perceptions and SI. CONCLUSION: This intensive longitudinal study illustrates momentary interpersonal signatures of an emerging suicidal crisis. Among people with BPD at high risk for suicide, interpersonal triggers initiate a cascade of affective dysregulation, which in turn gives rise to SI.

Neural connectivity biotypes: associations with internalizing problems throughout adolescence.

BACKGROUND: Neurophysiological patterns may distinguish which youth are at risk for the well-documented increase in internalizing symptoms during adolescence. Adolescents with internalizing problems exhibit altered resting-state functional connectivity (RSFC) of brain regions involved in socio-affective processing. Whether connectivity-based biotypes differentiate adolescents' levels of internalizing problems remains unknown. METHOD: Sixty-eight adolescents (37 females) reported on their internalizing problems at ages 14, 16, and 18 years. A resting-state functional neuroimaging scan was collected at age 16. Time-series data of 15 internalizing-relevant brain regions were entered into the Subgroup-Group Iterative Multi-Model Estimation program to identify subgroups based on RSFC maps. Associations between internalizing problems and connectivity-based biotypes were tested with regression analyses. RESULTS: Two connectivity-based biotypes were found: a Diffusely-connected biotype (N = 46), with long-range fronto-parietal paths, and a Hyper-connected biotype (N = 22), with paths between subcortical and medial frontal areas (e.g. affective and default-mode network regions). Higher levels of past (age 14) internalizing problems predicted a greater likelihood of belonging to the Hyper-connected biotype at age 16. The Hyper-connected biotype showed higher levels of concurrent problems (age 16) and future (age 18) internalizing problems. CONCLUSIONS: Differential patterns of RSFC among socio-affective brain regions were predicted by earlier internalizing problems and predicted future internalizing problems in adolescence. Measuring connectivity-based biotypes in adolescence may offer insight into which youth face an elevated risk for internalizing disorders during this critical developmental period.

The Hierarchical Taxonomy of Psychopathology (HiTOP): A Quantitative Nosology Based on Consensus of Evidence.

Traditional diagnostic systems went beyond empirical evidence on the structure of mental health. Consequently, these diagnoses do not depict psychopathology accurately, and their validity in research and utility in clinicalpractice are therefore limited. The Hierarchical Taxonomy of Psychopathology (HiTOP) consortium proposed a model based on structural evidence. It addresses problems of diagnostic heterogeneity, comorbidity, and unreliability. We review the HiTOP model, supporting evidence, and conceptualization of psychopathology in this hierarchical dimensional framework. The system is not yet comprehensive, and we describe the processes for improving and expanding it. We summarize data on the ability of HiTOP to predict and explain etiology (genetic, environmental, and neurobiological), risk factors, outcomes, and treatment response. We describe progress in the development of HiTOP-based measures and in clinical implementation of the system. Finally, we review outstanding challenges and the research agenda. HiTOP is of practical utility already, and its ongoing development will produce a transformative map of psychopathology.

Disentangling cognitive processes in externalizing psychopathology using drift diffusion modeling: Antagonism, but not disinhibition, is associated with poor cognitive control.

Although externalizing psychopathology has been linked to deficits in cognitive control, the cognitive processes underlying this association are unclear. Here, we provide a theoretical account of how research on cognitive processes can help to integrate and distinguish personality and psychopathology. We then apply this account to connect the two major subcomponents of externalizing, Antagonism and Disinhibition, with specific control processes using a battery of inhibitory control tasks and corresponding computational modeling. Participants (final N = 104) completed the flanker, go/no-go, and recent probes tasks, as well as normal and maladaptive personality inventories and measures of psychological distress. We fit participants' task behavior using a hierarchical drift diffusion model (DDM) to decompose their responses into specific cognitive processes. Using multilevel structural equation models, we found that Antagonism was associated with faster RTs on the flanker task and lower accuracy on flanker and go/no-go tasks. These results were complemented by DDM parameter associations: Antagonism was linked to decreased threshold and drift rate parameter estimates in the flanker task and a decreased drift rate on no-go trials. Altogether, our findings indicate that Antagonism is associated with specific impairments in fast (sub-second) inhibitory control processes involved in withholding prepared/prepotent responses and filtering distracting information. Disinhibition and momentary distress, however, were not associated with task performance.

Problems with Centrality Measures in Psychopathology Symptom Networks: Why Network Psychometrics Cannot Escape Psychometric Theory.

Understanding patterns of symptom co-occurrence is one of the most difficult challenges in psychopathology research. Do symptoms co-occur because of a latent factor, or might they directly and causally influence one another? Motivated by such questions, there has been a surge of interest in network analyses that emphasize the putatively direct role symptoms play in influencing each other. In this critical paper, we highlight conceptual and statistical problems with using centrality measures in cross-sectional networks. In particular, common network analyses assume that there are no unmodeled latent variables that confound symptom co-occurrence. The traditions of clinical taxonomy and test development in psychometric theory, however, greatly increase the possibility that latent variables exist in symptom data. In simulations that include latent variables, we demonstrate that closeness and betweenness are vulnerable to spurious covariance among symptoms that connect subgraphs (e.g., diagnoses). We further show that strength is redundant with factor loading in several cases. Finally, if a symptom reflects multiple latent causes, centrality metrics reflect a weighted combination, undermining their interpretability in empirical data. Our results suggest that it is essential for network psychometric approaches to examine the evidence for latent variables prior to analyzing or interpreting patterns at the symptom level. Failing to do so risks identifying spurious relationships or failing to detect causally important effects. Altogether, we argue that centrality measures do not provide solid ground for understanding the structure of psychopathology when latent confounding exists.

Predicting future courses of psychotherapy within a grouped LASSO framework.

Objective: There is a paucity of studies examining the experience of clients who undergo multiple courses of psychotherapy. Conducted within a large practice research network, this study demonstrated that returning therapy clients comprise a considerable portion of the clinical population in university counseling settings, and identified variables associated with return to therapy. Method: Utilizing data spanning 2013 to 2017, statistical variable selection for predicting return to therapy was conducted via grouped least absolute shrinkage and selection operator (grouped LASSO) applied to logistic regression. The grouped LASSO approach is described in detail to facilitate learning and replication. The paper also addresses methodological considerations related to this approach, such as sample size, generalizability, as well as general strengths and limitations. Results: Attendance rate, duration of initial treatment course, social anxiety, perceived social support, academic distress, and alcohol use were identified as predictive of return to therapy. Conclusions: Findings could help inform more cost-effective policies for session limits (e.g., extending session limits for clients with social anxiety), referral decisions (e.g., for clients with alcohol use problems), and appointment reminders (based on the association between poor attendance rate and return to therapy). Taking into account the many reasons that can explain why clients do or do not return to therapy, these findings also could inform clinicians' early case conceptualizations and treatment interventions.

Positive impact of genetic counseling assistants on genetic counseling efficiency, patient volume, and cost in a cancer genetics clinic.

PURPOSE: Cancer genetics clinics have seen increasing demand, challenging genetic counselors (GCs) to increase efficiency and prompting some clinics to implement genetic counseling assistants (GCAs). To evaluate the impact of GCAs on Geisinger's cancer genetics clinic, we tracked GC time utilization, new patient volume, and clinic cost per patient before and after implementing a GCA program. METHODS: GCs used time-tracking software while completing preappointment activities. Electronic health records were reviewed for appointment length and number of patients per week. Internal salary data for GCs and GCAs were used to calculate clinic costs per patient. RESULTS: Time spent by GCs completing each preappointment activity (21.8 vs. 15.1 minutes) and appointment length (51.6 vs. 44.5 minutes) significantly decreased after GCA program implementation (p values < 0.001). New patients per week per GC significantly increased (7.9 vs. 11.4, p < 0.001). Weekly clinic cost per patient significantly decreased ($233 vs. $176, p = 0.03). CONCLUSION: Implementing a GCA program increased GC efficiency in preappointment activities and clinic appointments, increased patient volume, and decreased clinic cost per patient. Such a program can improve access to GC services and assist GCs in focusing on the direct patient care for which they are specially trained.

Disinhibition and Detachment in Adolescence: A Developmental Cognitive Neuroscience Perspective on the Alternative Model for Personality Disorders.

Personality pathology often emerges during adolescence, but attempts to understand its neurocognitive basis have traditionally been undermined by problems associated with the categorical classification of personality disorders. In contrast, dimensional models of personality pathology, such as the Alternative Model for Personality Disorders (AMPD) in DSM-5, may provide a stronger foundation for neurobiological investigations of maladaptive individual differences in personality. As an example, we review studies of the adolescent development of reward processing and cognitive control and connect these systems to the normal personality hierarchy and to two dimensions included in the AMPD - Detachment and Disinhibition. We argue that by linking developmental changes in these systems to the AMPD, researchers will be better positioned to understand the relationship between neurocognitive development and the expression of personality pathology in adolescence and early adulthood.

Neurocomputational mechanisms of adaptive learning in social exchanges.

Prior work on prosocial and self-serving behavior in human economic exchanges has shown that counterparts' high social reputations bias striatal reward signals and elicit cooperation, even when such cooperation is disadvantageous. This phenomenon suggests that the human striatum is modulated by the other's social value, which is insensitive to the individual's own choices to cooperate or defect. We tested an alternative hypothesis that, when people learn from their interactions with others, they encode prediction error updates with respect to their own policy. Under this policy update account striatal signals would reflect positive prediction errors when the individual's choices correctly anticipated not only the counterpart's cooperation but also defection. We examined behavior in three samples using reinforcement learning and model-free analyses and performed an fMRI study of striatal learning signals. In order to uncover the dynamics of goal-directed learning, we introduced reversals in the counterpart's behavior and provided counterfactual (would-be) feedback when the individual chose not to engage with the counterpart. Behavioral data and model-derived prediction error maps (in both whole-brain and a priori striatal region of interest analyses) supported the policy update model. Thus, as people continually adjust their rate of cooperation based on experience, their behavior and striatal learning signals reveal a self-centered instrumental process corresponding to reciprocal altruism.

Incentives facilitate developmental improvement in inhibitory control by modulating control-related networks.

Adolescence is a period of heightened sensitivity to incentives and relatively weak cognitive control, which may contribute to risky behaviors. Studies of brain activity have generally identified greater activation of the ventral striatum to rewards and less activation of prefrontal regions during control tasks in adolescents compared to adults. Little is known, however, about age-related changes in the functional brain networks underlying incentive processing and cognitive control. This cross-sectional study characterized the effects of incentives on inhibitory control during an oculomotor task using whole-brain functional connectivity analyses. During an fMRI scan, one hundred forty typically developing individuals completed an incentivized antisaccade task consisting of incentive cue, preparation, and response phases. We found that task modulation of control networks increased gradually from childhood to adulthood, whereas a network including ventral striatum and ventromedial prefrontal cortex displayed an adolescent-specific peak in response to the receipt of outcomes, consistent with dual-systems models. Notably, however, greater modulation of salience and motor networks during the preparation phase mediated age-related improvements in antisaccade accuracy, whereas adolescent enhancement of value-related circuitry did not. Relative to neutral cues, both reward and loss cues enhanced task-related connectivity of the salience network when preparing to inhibit a saccade. Altogether, our findings suggest that incentives facilitate inhibitory control by enhancing the salience of one's responses and that over development, the recruitment of functional networks involved in saliency and motor preparation supports better performance.

The Contribution of Network Organization and Integration to the Development of Cognitive Control.

Cognitive control, which continues to mature throughout adolescence, is supported by the ability for well-defined organized brain networks to flexibly integrate information. However, the development of intrinsic brain network organization and its relationship to observed improvements in cognitive control are not well understood. In the present study, we used resting state functional magnetic resonance imaging (RS-fMRI), graph theory, the antisaccade task, and rigorous head motion control to characterize and relate developmental changes in network organization, connectivity strength, and integration to inhibitory control development. Subjects were 192 10-26-y-olds who were imaged during 5 min of rest. In contrast to initial studies, our results indicate that network organization is stable throughout adolescence. However, cross-network integration, predominantly of the cingulo-opercular/salience network, increased with age. Importantly, this increased integration of the cingulo-opercular/salience network significantly moderated the robust effect of age on the latency to initiate a correct inhibitory control response. These results provide compelling evidence that the transition to adult-level inhibitory control is dependent upon the refinement and strengthening of integration between specialized networks. Our findings support a novel, two-stage model of neural development, in which networks stabilize prior to adolescence and subsequently increase their integration to support the cross-domain incorporation of information processing critical for mature cognitive control.

Functional connectivity differences in autism during face and car recognition: underconnectivity and atypical age-related changes.

Face recognition abilities improve between adolescence and adulthood over typical development (TD), but plateau in autism, leading to increasing face recognition deficits in autism later in life. Developmental differences between autism and TD may reflect changes between neural systems involved in the development of face encoding and recognition. Here, we focused on whole-brain connectivity with the fusiform face area (FFA), a well-established face-preferential brain region. Older children, adolescents, and adults with and without autism completed the Cambridge Face Memory Test, and a matched car memory test, during fMRI scanning. We then examined task-based functional connectivity between the FFA and the rest of the brain, comparing autism and TD groups during encoding and recognition of face and car stimuli. The autism group exhibited underconnectivity, relative to the TD group, between the FFA and frontal and primary visual cortices, independent of age. Underconnectivity with the medial and rostral lateral prefrontal cortex was face-specific during encoding and recognition, respectively. Conversely, underconnectivity with the L orbitofrontal cortex was evident for both face and car encoding. Atypical age-related changes in connectivity emerged between the FFA and the R temporoparietal junction, and R dorsal striatum for face stimuli only. Similar differences in age-related changes in autism emerged for FFA connectivity with the amygdala across both face and car recognition. Thus, underconnectivity and atypical development of functional connectivity may lead to a less optimal face-processing network in the context of increasing general and social cognitive deficits in autism.

Protracted development of executive and mnemonic brain systems underlying working memory in adolescence: A longitudinal fMRI study.

Working memory (WM), the ability to hold information on-line to guide planned behavior, improves through adolescence in parallel with continued maturation of critical brain systems supporting cognitive control. Initial developmental neuroimaging studies with one or two timepoints have provided important though varied results limiting our understanding of which and how neural systems change during this transition into mature WM. In this study, we leverage functional magnetic resonance imaging (fMRI) longitudinal data spanning up to 9 years in 129 normally developing individuals to identify which systems demonstrate growth changes that accompany improvements in WM performance. We used a memory guided saccade task that allowed us to probe encoding, pure maintenance, and retrieval neural processes of WM. Consistent with prior research, we found that WM performance continued to improve into the early 20's. fMRI region of interest (ROI) analyses revealed developmental (1) increases in sensorimotor-related (encoding/retrieval) activity in visual cortex from childhood through early adulthood that were associated with WM accuracy and (2) decreases in sustained (maintenance) activity in executive regions from childhood through mid-adolescence that were associated with response latency in childhood and early adolescence. Together these results provide compelling evidence that underlying the maturation of WM is a transition from reliance on executive systems to specialized regions related to the domain of mnemonic requirements of the task leading to optimal performance.

Modulation of reward-related neural activation on sensation seeking across development.

Sensation seeking is a personality construct associated with an increased propensity for engaging in risk-taking. Associations with deleterious outcomes ranging from mental health impairments to increased mortality rates highlight important public health concerns related to this construct. Although some have suggested that increased neural responsivity to reward within the ventral striatum (e.g., nucleus accumbens) may drive sensation seeking behaviors, few studies have examined the neural mechanisms associated with stable individual differences in sensation seeking across development. To address this issue, the current study used functional magnetic resonance imaging to examine the association between neural responding to reward and stable patterns of sensation seeking across a three-year follow-up period among healthy adolescents and young adults (N = 139). Results indicated that during early adolescence (~ages 10-12), increased reactivity to reward within the nucleus accumbens (NAcc) was associated with lower levels of sensation seeking across a three-year follow-up. In middle adolescence (~ages 12-16), there was no evidence of a relationship between NAcc reactivity and sensation seeking. However, during the transition from late adolescence into adulthood (~ages 17-25), heightened reward-related reactivity in the NAcc was linked to increased sensation seeking. Findings suggest that the neural mechanisms underlying individual differences in trait-like levels of sensation seeking change from early to late adolescence.

Impulsive traits and unplanned suicide attempts predict exaggerated prefrontal response to angry faces in the elderly.

OBJECTIVE: Abnormal responses to social stimuli are seen in people vulnerable to suicidal behavior, indicating possible disruptions in the neural circuitry mediating the interpretation of socioemotional cues. These disruptions have not been empirically related to psychological and cognitive pathways to suicide. In the present study of older suicide attempters, we examined neural responses to emotional faces and their relationship to impulsivity, one of the components of the suicidal diathesis. METHODS: Using functional magnetic resonance imaging, we recorded neurohemodynamic responses to angry faces in a carefully characterized sample of 18 depressed elderly with history of suicide attempts, 13 depressed nonsuicidal patients, and 18 healthy individuals, all aged 60+. Impulsivity was assessed with the Social Problem Solving Inventory Impulsivity/Carelessness Style subscale and Barratt Impulsiveness Scale. The Suicide Intent Scale planning subscale was used to describe the degree of planning associated with the most lethal attempt. RESULTS: Depression and history of attempted suicide were not associated with neural responses to angry faces, failing to replicate earlier studies. Higher impulsivity, however, predicted exaggerated responses to angry faces in fronto-opercular and dorsomedial prefrontal cortex (pcorr <0.05). Poorly planned suicide attempts also predicted increased fronto-opercular responses. Results were robust to effects of medication exposure, comorbid anxiety and addiction, severity of depression, burden of physical illness, and possible brain injury from suicide attempts. CONCLUSION: Impulsive traits and history of unplanned suicide attempts partly explain the heterogeneity in neural responses to angry faces in depressed elderly. Displays of social emotion command excessive cortical processing in impulsive suicide attempters.

Developmental stages and sex differences of white matter and behavioral development through adolescence: a longitudinal diffusion tensor imaging (DTI) study.

White matter (WM) continues to mature through adolescence in parallel with gains in cognitive ability. To date, developmental changes in human WM microstructure have been inferred using analyses of cross-sectional or two time-point follow-up studies, limiting our understanding of individual developmental trajectories. The aims of the present longitudinal study were to characterize the timing of WM growth and investigate how sex and behavior are associated with different developmental trajectories. We utilized diffusion tensor imaging (DTI) in 128 individuals aged 8-28, who received annual scans for up to 5 years and completed motor and cognitive tasks. Flexible nonlinear growth curves indicated a hierarchical pattern of WM development. By late childhood, posterior cortical-subcortical connections were similar to adults. During adolescence, WM microstructure reached adult levels, including frontocortical, frontosubcortical and cerebellar connections. Later to mature in adulthood were major corticolimbic association tracts and connections at terminal gray matter sites in cortical and basal ganglia regions. These patterns may reflect adolescent maturation of frontal connectivity supporting cognitive abilities, particularly the protracted refinement of corticolimbic connectivity underlying cognition-emotion interactions. Sex and behavior also played a large role. Males showed continuous WM growth from childhood through early adulthood, whereas females mainly showed growth during mid-adolescence. Further, earlier WM growth in adolescence was associated with faster and more efficient responding and better inhibitory control whereas later growth in adulthood was associated with poorer performance, suggesting that the timing of WM growth is important for cognitive development.