Poster #11 - The neural mechanisms supporting habituation to audio-visual associations in infants: An fNIRS and looking time study

• In Event: 1-061 - Poster Session 02
  In Poster Session: PS 02 Section - Attention, Learning, Memory

Integrative Statement

Habituation serves as a key method in both facilitating and assessing learning in infants and young children, yet its underlying neural mechanism has yet to be directly investigated. Since its first usage (Fantz, 1964), at least five decades of developmental research have been built on this paradigm. While the nature of habituation has been theorized about at length (e.g., Hunter & Ames, 1988; Kidd, Piantadosi, & Aslin, 2012), examining the underlying neural changes of this behavior can shed light on the learning mechanisms supporting habituation and how these learning processes vary (e.g., across tasks, age, population).
Previous neuroimaging studies found that the hemodynamic response during infant learning of audiovisual stimuli is defined by an increase followed by a decrease in activity (Kersey & Emberson, 2016), and that different cortical regions may show unique patterns of activation in response to familiar or novel stimuli (Nakano, Watanabe, Homae, & Taga, 2009). However, these studies did not measure the entire course of infants’ habituation as the length of each stimulus was fixed in advance and did not depend on individual babies’ looking times. Moreover, no fNIRS study has integrated infant behavior (as measured by looking time) with neural responses. Integrating learning outcomes, as measured behaviorally, with neural responses in infants will facilitate the integration of the neural signatures of learning and the decades of behavioral investigations of these phenomena.
Using fNIRS (Aslin, Shukla, & Emberson, 2015), we now investigate the neural correlate of habituation in infants, as they go through a well-established protocol for audio-visual association learning (Stager & Werker, 1997). Crucially, stimulus presentation is presented contingent on infant looking behavior and looking time measures of learning are collected after habituation. Infants were habituated to repeated presentation of patterned movie clips with arbitrary audio-visual association, while their cortical responses were recorded using a whole-head fNIRS system (74 channels). Each session consisted of habituation phase and a test phase. In the habituation phase, infants watched two paired audio-visual stimuli (A1V1 and A2V2) that were presented until the infant looked away for longer than 1 second. When the infant looked back at the screen for longer than 3 seconds, another paired audio-visual stimulus was presented (either the same A1V1 pair or the other A2V2 pair). The habituation phase concluded when the average looking time for the last three presentation was shorter that half of the average looking time for the first three presentations. At the test phase, infants were presented with familiar pairs (A1V1 and A2V2) and novel pairs (A1V2 and A2V1) and their looking time was measured.
So far, 16 out of 26 infants completed both phases of the experiment and showed habituation to the presented stimuli. Preliminary results suggest that multiple cortical locations exhibit attenuation where the response to the repeated stimuli was attenuated from presentation to presentation. These findings not only shed light on the underlying neural mechanism of habituation but also represented the first integration of neural measures with behavioral measures in young infants. This integration of methods is crucial to understanding the mechanisms supporting behavior and behavioral change in early development.

Authors

• Alice F Wang, Princeton University
  Presenting Author

• Sagi Jaffe-Dax, Princeton University
  Non-Presenting Author

• Lauren Emberson, Princeton University
  Non-Presenting Author