The right hemispheric dominance for face perception in preschool children depends on the visual discrimination level

Abstract

The developmental origin of human adults’ right hemispheric dominance in response to face stimuli remains unclear, in particular because young infants’ right hemispheric advantage in face‐selective response is no longer present in preschool children, before written language acquisition. Here we used fast periodic visual stimulation (FPVS) with scalp electroencephalography (EEG) to test 52 preschool children (5.5 years old) at two different levels of face discrimination: discrimination of faces against objects, measuring face‐selectivity, or discrimination between individual faces. While the contrast between faces and nonface objects elicits strictly bilateral occipital responses in children, strengthening previous observations, discrimination of individual faces in the same children reveals a strong right hemispheric lateralization over the occipitotemporal cortex. Picture‐plane inversion of the face stimuli significantly decreases the individual discrimination response, although to a much smaller extent than in older children and adults tested with the same paradigm. However, there is only a nonsignificant trend for a decrease in right hemispheric lateralization with inversion. There is no relationship between the right hemispheric lateralization in individual face discrimination and preschool levels of readings abilities. The observed difference in the right hemispheric lateralization obtained in the same population of children with two different paradigms measuring neural responses to faces indicates that the level of visual discrimination is a key factor to consider when making inferences about the development of hemispheric lateralization of face perception in the human brain.

Understanding human individuation of unfamiliar faces with oddball fast periodic visual stimulation and electroencephalography

“…For biophysical reasons, young children tend to have ERP components of particularly high amplitudes, which decrease progressively with age (e.g., Kuefner et al., 2010). This also appears to be the case for the FI response (e.g., close to 1.5 µV over occipito‐temporal channels in 5 year‐olds, Lochy et al, 2020; Figure 18), which is roughly of the same amplitude as in adults. However, their stimulus presentation response is also much larger (2.1 µV in adults, but a response reaching up to 3 µV in young children, Figure 5 and Lochy et al, 2020).…”

“…In five‐year‐old children, a clear FI response at F / n is present over the right occipito‐temporal cortex, as for adults (Lochy et al, 2020; Figure 18). Note that a clear right lateralization is present here before formal reading acquisition, thus contradicting the hypothesis that the right lateralization for faces results from a competition with letter and word representations during formal reading acquisition (Behrmann & Plaut, 2013; Dehaene, Cohen, Morais, & Kolinsky, 2015).…”

“…This harmonic amplitude distribution is somewhat variable across individuals and relatively stable within individuals, providing an additional marker of the FI function in individuals (Dzhelyova et al, 2019; see Section 7). Another interesting example is provided by developmental studies: as noted above, the bulk of the FI response (about 60%) in this paradigm is accounted for by the first harmonic in 5‐year‐old children but only 18% in adults, providing an additional marker of the developmental trajectory of FI (Lochy et al, 2020; see Section 7.4).…”

“…First, while children's FI response amplitude is quite high (similarly as with ERPs in children, e.g., Kuefner, de Heering, Jacques, Palmero‐Soler, & Rossion, 2010), it is much lower in SNR (amplitude relative to background EEG noise) than for adults, and its amplitude relative to that of the general visual response at F is smaller than in adults. Second, for children, the first harmonic of the oddball frequency ( F / n ) accounts for the bulk of the FI response (about 60%, Figure 18a, compared to only 18% in adults, Figure 4), suggesting a simpler or longer response (Lochy et al, 2020 see Box 1). Third, strikingly, there is almost no reduction of the FI response with picture‐plane inversion in children (i.e., a mere 11% amplitude decrease, just reaching significance in a large sample; Lochy et al, 2020).…”

“…Second, for children, the first harmonic of the oddball frequency ( F / n ) accounts for the bulk of the FI response (about 60%, Figure 18a, compared to only 18% in adults, Figure 4), suggesting a simpler or longer response (Lochy et al, 2020 see Box 1). Third, strikingly, there is almost no reduction of the FI response with picture‐plane inversion in children (i.e., a mere 11% amplitude decrease, just reaching significance in a large sample; Lochy et al, 2020). Interestingly, in an older population of 8‐ to 12‐year‐olds, the EEG responses are much more adult‐like, with a large and complex FI response distributed over many harmonic frequencies, and a large inversion effect (about 50%; Figure 18b; Vettori, Dzhelyova, et al, 2019).…”