Search
Browse By Day
Browse By Time
Browse By Panel
Browse By Session Type
Browse By Topic Area
Search Tips
Register for SRCD21
Personal Schedule
Change Preferences / Time Zone
Sign In
X (Twitter)
Does neural representation in the visual cortex change as we develop? In the adult brain, sensory inputs relevant to context or goals are often better represented in the brain at the expense of task-irrelevant information (Kamitani & Tong, 2006; Jehee et al., 2011). This attentional enhancement of sensory responses in the visual cortex is driven by top-down signals from the associative cortex, such as the prefrontal and parietal regions (Gazzaley & Nobre, 2012; Miller & D’Esposito, 2005; Scolari, Seidl-Rathkopf & Kastner, 2015). However, the associative cortex undergoes massive developmental shifts both anatomically (Sowell et al., 2004) and functionally (Bunge et al., 2002; Finn et al., 2018) from infancy to late childhood, suggesting that the top-down influences on the visual cortex may also change with development. Further, research has shown that the connectivity across brain regions, especially long-range connectivity, matures gradually over the course of development (de Bie et al., 2011; Supekar et al., 2010). Given these findings, it is possible that neural representations in the child visual cortex might not be modulated by top-down influences to the same extent as in adults.
In the present study, we therefore ask how neural representations in the child visual cortex is modulated by top-down signals. We measured brain activity using fMRI while adults (21-31 years) and children (7-10 years) performed a one-back task in which they were directed to attend to either the motion or the object present in a complex display where both objects and motion were always present. In other words, in the motion-attended condition, participants performed the one-back task on motion while ignoring the object, and vice versa. To examine how these features are represented in the child and adult brain, we used the multivoxel pattern analysis and compared decoding accuracy of task-relevant and task-irrelevant features.
We found that behaviorally, adults and children performed the one-back task equally well (Fig A). However, there were differences in the neural representations in adults’ and children’s visual cortices. In the adult visual cortex, replicating the previous literature, we found higher decoding accuracies for task-relevant features (i.e., object categories in the object-attended condition; see Fig B) compared to those for task-irrelevant features (i.e., objects in the motion-attended condition). However, in the child visual cortex, both task-relevant and task-irrelevant features are equally well decoded beyond the chance, which demonstrates less top-down modulation of visual cortex in children. These findings show that developing visual cortex appears to represent more information than the adult visual cortex, which is known to prioritize task-related information. Also, this research suggests that unlike measures of brain structure, the functional properties of the visual cortex continue to develop in the childhood.