RESUMO
BACKGROUND: Puberty depicts a period of profound and multifactorial changes ranging from social to biological factors. While brain development in youths has been studied mostly from an age perspective, recent evidence suggests that pubertal measures may be more sensitive to study adolescent neurodevelopment, however, studies on pubertal timing in relation to brain development are still scarce. METHODS: We investigated if pre- vs. post-menarche status can be classified using machine learning on cortical and subcortical structural magnetic resonance imaging (MRI) data from strictly age-matched adolescent females from the Adolescent Brain Cognitive Development (ABCD) cohort. For comparison of the identified menarche-related patterns to age-related patterns of neurodevelopment, we trained a brain age prediction model on data from the Philadelphia Neurodevelopmental Cohort and applied it to the same ABCD data, yielding differences between predicted and chronological age referred to as brain age gaps. We tested the sensitivity of both these frameworks to measures of pubertal maturation, specifically age at menarche and puberty status. RESULTS: The machine learning model achieved moderate but statistically significant accuracy in the menarche classification task, yielding for each subject a class probability ranging from 0 (pre-) to 1 (post- menarche). Comparison to brain age predictions revealed shared and distinct patterns of neurodevelopment captured by both approaches. Continuous menarche class probabilities were positively associated with brain age gaps, but only the menarche class probabilities-not the brain age gaps-were associated with age at menarche. CONCLUSIONS: This study demonstrates the use of a machine learning model to classify menarche status from structural MRI data while accounting for age-related neurodevelopment. Given its sensitivity towards measures of puberty timing, our work suggests that menarche class probabilities may be developed toward an objective brain-based marker of pubertal development.
Puberty is a period of substantial changes in the life of youths, and these include profound brain changes. Most studies have investigated age related changes in brain development, recent work however suggests that looking at brain development through the lens of pubertal development can provide additional insights beyond age effects. We here analyzed brain imaging data from a group of same-aged adolescent girls from the Adolescent Brain Cognitive Development study. Our goal was to investigate if we could determine from brain images whether a girl had started her menstrual period (menarche) or not, and we used machine learning to classify between them. This machine learning model does not just return a "yes/no" decision, but also returns a number between 0 and 1 indicating a probability to be pre- (0) or post- (1) menarche. To rule out that our approach only maps age-related development, we selected a strictly age-matched sample of girls and compared our classification model to a brain age model trained on independent individuals. Our model classified between pre- and post-menarche with moderate accuracy. The obtained class probability was partly related to age-related brain development, but only the probability was significantly associated with pubertal timing (age at menarche). In summary, our study uses a machine learning model to estimate whether a girl has reached menarche based on her brain structure. This approach offers new insights into the connection between puberty and brain development and might serve as an objective way to assess pubertal timing from imaging data.
Assuntos
Menarca , Puberdade , Adolescente , Humanos , Feminino , EncéfaloRESUMO
Visual attention is mainly goal directed and allocated based on the upcoming action. However, it is unclear how far this feature of gaze behaviour generalizes in more naturalistic settings. The present study investigates the influence of action affordances on active inference processes revealed by eye movements during interaction with familiar and novel tools. In a between-subject design, a cohort of participants interacted with a virtual reality controller in a low-realism environment; another performed the task with an interaction setup that allowed differentiated hand and finger movements in a high-realism environment. We investigated the differences in odds of fixations and their eccentricity towards the tool parts before action initiation. The results show that participants fixate more on the tool's effector part before action initiation when asked to produce tool-specific movements, especially with unfamiliar tools. These findings suggest that fixations are made in a task-oriented way to plan the distal goals of producing the task- and tool-specific actions well before action initiation. Moreover, with more realistic action affordance, fixations were biased towards the tool handle when it was oriented incongruent with the subjects' handedness. We hypothesize that these fixations are made towards the proximal goal of planning the grasp even though the perceived action on the tools is identical for both experimental setups. Taken together, proximal and distal goal-oriented planning is contextualized to the realism of action/interaction afforded by an environment.