Presentation

Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Poster Slams

What is special about (misspelled) logotypes? Examining the brain signature of misspelled logos

Poster B21 in Poster Session B and Reception, Thursday, October 6, 6:30 - 8:30 pm EDT, Millennium Hall

Maria Fernandez-Lopez¹, Francisco Rocabado², Manuel Perea^1,2, Melanie Labusch², Ana Marcet¹, Marta Vergara-Martinez¹; ¹University of València, ²Nebrija University

Companies and products are identified by their brand names, typically presented as logotypes (i.e., with a given font style, color, and design). While this graphic information provides a distinctive image that facilitates recognition, it also makes them particularly vulnerable to counterfeiting through misspelled branding (e.g., amzaon). Indeed, they are affected by letter transpositions to a larger degree than common misspelled words (e.g., JUGDE) (see Perea et al., 2021, 2022, for behavioral evidence). Why are logotypes especially susceptible to misspellings? While common words are defined by their orthographic characteristics in written language, logotypes are defined by both their graphic and orthographic characteristics. In this line, neurally-inspired models of visual (common) word recognition (e.g., Dehaene et al., 2005; Grainger et al., 2008) assume that information such as color or lettering style is lost early during the processing of common words. The underlying logic is that word recognition is based on retrieving abstract orthographic characteristics resulting from an explicit learning process. However, logotype identification can be particularly dependent on color and lettering style. As a result, orthographic encoding would play a secondary role, thus making logotypes easily susceptible to letter transposition effects (e.g., Perea et al., 2021). The goal of the present ERP experiment was to uncover the processing of letter position information when recognizing logotypes. Participants were presented with correctly spelled (SAMSUNG) or incorrectly spelled (letter-transposition: SASMUNG; letter-replacement: SARVUNG) logotypes in a semantic categorization task. Participants had to decide whether the brand name’s logotype was related to travel or not (e.g., RYANAIR [a low-cost European carrier] vs. SAMSUNG)—the focus was on the no-go responses (i.e., non-travel logos). The ERP results showed higher amplitude for the replaced-letter brand names than for the transposed-letter and correctly-spelled brand names at an early stage of processing (P200). The difference between the replaced-letter vs. transposed-letter and correctly-spelled brand names remained until 600ms. Critically, there were no differences in amplitude between the transposed letter and the correctly-spelled brand names over most scalp electrodes. This pattern of results critically differs from previous results on transposition effects with common words. In a semantic categorization task, Vergara-Martínez et al. (2013) found amplitude differences at around 300ms between the correctly-spelled words (JUDGE) and both the replaced (JUPTE) and transposed (JUGDE) pseudowords. While these differences remained between the replaced-letter and correctly-spelled words until 600ms, they vanished for the transposed and correctly-spelled words between 360-470ms. Notably, at a later window (470-580 ms), both types of pseudowords elicited larger negativities than words. In summary, the present findings place limits on a tenet of the leading word recognition models about an initial confusion in orthographic processing (e.g., letter position encoding) that is eventually resolved. At least for logotypes, this processing does not seem to reach the last orthographic verification stage. Instead, information such as color or lettering style can be more relevant for logotype identification than letter position. Critically, these results challenge those theoretical models, which predict a similar time course in the processing of letter position coding for misspelled logotypes and misspelled common words.

Topic Areas: Reading, Writing and Spelling