Is the rapid adaptation paradigm too rapid? Implications for face and object processing

Abstract

Rapid adaptation is an adaptation procedure in which adaptors and test stimuli are presented in rapid succession. The current study tested the validity of this method for early ERP components by investigating the specificity of the adaptation effect on the face-sensitive N170 ERP component across multiple test stimuli. Experiments 1 and 2 showed identical response patterns for house and upright face test stimuli using the same adaptor stimuli. The results were also identical to those reported in a previous study using inverted face test stimuli (Nemrodov and Itier, 2011). In Experiment 3 all possible adaptor-test combinations between upright face, house, chair and car stimuli were used and no interaction between adaptor and test category, expected in the case of test-specific adaptation, was found. These results demonstrate that the rapid adaptation paradigm does not produce category-specific adaptation effects around 170-200 ms following test stimulus onset, a necessary condition for the interpretation of adaptation results. These results suggest the rapid categorical adaptation paradigm does not work.

When rapid adaptation paradigm is not too rapid: Evidence of face-sensitive N170 adaptation effects

“…We first replicated recent findings that face adaptors induced N170 adaptation effects for both face and house tests in the short SOA (Nemrodov & Itier, 2012). Those category-general effects are presumably attributed to the contamination of ERPs evoked by temporally adjacent adaptors and tests (Luck, 2005;Woldorff, 1993), implicating adaptor-locked interferences on early ERPs of tests (e.g., P1 and N170).…”

“…For both P1 and N170 components, the baseline-to-peak measurement was used in order to compare the current results with those of previous studies (e.g., Eimer et al, 2010;Nemrodov & Itier, 2012). Importantly, the baseline-to-peak measurement of N170 amplitude was supplemented with a peak-to-peak measurement (N170-P1 amplitude), in which baseline-to-peak P1 amplitude was subtracted from baselineto-peak N170 amplitude (Goffaux, Gauthier, & Rossion, 2003;Itier & Taylor, 2002;Rossion, Joyce, Cottrell, & Tarr, 2003).…”

“…For instance, face adaptors, when compared with non-face adaptors, induced larger N170 adaptation effect for face tests, but it remains unknown whether such an adaptation advantage induced by face adaptors is specific to face tests or general to non-face tests. In the study of Nemrodov and Itier (2012), the authors observed the N170 adaptation advantage induced by face adaptors for both face and non-face tests, demonstrating a category-general N170 adaptation effect at test level. These findings led to the notion that the rapid categorical adaptation paradigm may not serve as a suitable tool to tag the sensitivity of neural activity.…”

“…Against the back drop of these promising applications of N170 adaptation effect is recent evidence for adaptor-locked interferences in the rapid adaptation paradigm (Nemrodov & Itier, 2012). The authors argued that only one stimulus category (e.g., faces) was used as tests in most studies employing N170 adaptation effect, leaving the possibility that adaptation effect induced by an adaptor category is not specific to the following test category, but rather a general effect to other test categories.…”

“…The category-general N170 adaptation effect (at the test level) is presumably caused by adaptor-locked interferences such as late neural responses and offset effects evoked by S1 (Nemrodov & Itier, 2012), regarding that the SOA between S1 and S2 is short (400-500 ms) and un-jittered in most studies. Hence, there is an urgent need to systematically manipulate SOA parameters between S1 and S2 in order to reveal the extent to which adaptor-locked interferences confound N170 adaptation effect in different SOA parameters and to examine a suitable SOA parameter for future studies to use.…”