Does social distancing affect the processing of brand logos?

Abstract Social distancing and isolation have been imposed to contrast the spread of COVID‐19. The present study investigates whether social distancing affects our cognitive system, in particular the processing of different types of brand logos in different moments of the pandemic spread in Italy. In a size discrimination task, six different logos belonging to three categories (letters, symbols, and social images) were presented in their original format and spaced. Two samples of participants were tested: one just after the pandemic spread in Italy, the other one after 6 months. Results showed an overall distancing effect (i.e., spaced stimuli are processed slower than original ones) that interacted with the sample, revealing a significant effect only for participants belonging to the second sample. However, both groups showed a distancing effect modulated by the type of logo as it only emerged for social images. Results suggest that social distancing behaviors have been integrated in our cognitive system as they appear to affect our perception of distance when social images are involved.

distance and social interactions. For instance, Cartaud et al. (2020) investigated how wearing a facemask affects social distancing: participants reported a significant decrease in preferred interpersonal distance when the social interaction involved face stimuli wearing a facemask compared to face stimuli with no mask. Xu and Cheng (2021) observed that participants who were more risk averse practiced more social distancing and mask-wearing.
More generally, the influence of context on our cognitive processes has been well-documented in experimental psychology. For instance, in the Flanker task (Eriksen & Eriksen, 1974;Lappin et al., 2021 for recent discussion) participants are instructed to identify a relevant characteristic of the stimulus (e.g., the orientation of an arrow stimulus) that is presented in a context that can be congruent or incongruent with it. Thus, participants may need to identify an arrow pointing to the left among other arrows pointing to the left (congruent context), or among other arrows pointing to the right (incongruent context). Results demonstrate that participants are influenced by the context surrounding the target arrow as slower response times are observed with the incongruent compared to the congruent context, despite the fact that the latter is irrelevant to the task. Furthermore, research on memory highlighted the so-called priming effect, that is, the influence that the introduction of one stimulus (often a word or an image) has on how people respond to a subsequent stimulus (e.g., Collins & Loftus, 1975;Meyer & Schvaneveldt, 1971). For example, the word nurse is recognized faster following the word doctor than the word bread. More generally, priming occurs whenever exposure to one thing affects an individual's subsequent behavior or thoughts.
Moreover, in psycholinguistic research the importance of the context has been highlighted by the word superiority effect (WSE, Reicher, 1969;Wheeler, 1970) whereby letters are identified better when embedded within a word than in a nonword string, or when presented alone.
For the purposes of our study, it is also worth emphasizing that stimuli can be processed faster on the basis of their familiarity (e.g., Wang et al., 1994). In a visual search task, Wang et al. (1994) observed that when low-level features of targets and distractors were held constant, performance was affected by familiarity. Specifically, they obtained faster reaction times in a visual search with familiar stimuli (i.e., a single upright letter) compared to unfamiliar ones (i.e., a single inverted letter). Importantly, our object recognition system automatically connects perceived stimuli with stored knowledge (Grill-Spector & Kanwisher, 2005;Thorpe et al., 1996), leading to a faster recognition of stimuli we already know since semantic information is already available. In addition, there is evidence that we are able to detect an object faster when we are familiar with it (Konkle & Oliva, 2012).
On the basis of the above reported evidence highlighting the impact of COVID on our cognitive processes and social behaviors, the influence that context exerts on these processes, and how we typically react to familiar stimuli, we investigated how the current pandemic context impacts on the processing of original and spaced brand logos. This choice was motivated by an advertising campaign of some brands (e.g., Macdonald's, Coca-Cola), which have modified their logos to promote social distancing during the pandemic.
Logos as visual cues help firms communicate their identities and capture consumers' attention (Peracchio & Meyers-Levy, 2005). Given their importance, several studies have investigated the role of logo representation and, in particular, how the characteristics of logos can influence the perception of the properties of a product or a company.
For example, it has been demonstrated that preference judgments are influenced by the circularity or angularity of a logo (Jiang et al., 2016), its typographic character (Grohmann et al., 2013), its asymmetry (Luffarelli et al., 2019) and its orientation (Zhong et al., 2018; for a review see Kim & Lim, 2019). In addition, Philiastides and Ratcliff (2013) showed that logos and brands can have a great influence on decision-making processes as they can affect our choices if present on a specific clothing.
Given the role of logos in orienting consumers' behaviors, it could be relevant to investigate whether and to what extent their perception might be affected by social contextual constraints, such as those imposed by social distancing.
In the present study, we used a size discrimination task where participants had to determine whether the stimulus was large or small.
Original and spaced logos belonging to three different categories (letters, symbols, and social images) were used as targets. We expected to observe a distancing effect, whereby spaced logos, that is, unfamiliar logos, were processed slower than original ones, i.e., familiar logos, in line with previous evidence (Qin et al., 2014). Qin et al. (2014) indeed reported that in a visual search task familiar logos were found faster than unfamiliar ones. Moreover, we expected to observe that, within spaced logos, social images logos, evoking the practice of social distancing, might be processed slower than letters and symbols spaced logos because they are associated with forced and unnatural behavior.
This could result in a greater cognitive effort to elaborate them, which might be reflected by slower RT. Therefore, we predict a modulation of the distancing effect with respect to the different kinds of logo. More specifically, we hypothesize to find a greater distancing effect for social logo images as these images depict human-like figures, thus recalling an unnatural/unfamiliar social interaction such as the one imposed by anti-COVID measures. After all, social images recall a social dimension that is the aspect of our daily life most affected by the adopted behavior during the pandemic. In addition, we predict that the more one has been exposed to these measures, the greater the effect. To test this prediction, two different samples were tested: the first one just after the pandemic spread started in Italy and the population began acting following distancing practices (June to July 2020), and the second one after 6 months (January 2021), when the distancing behaviors were already well established. The distancing effect should be larger for participants who were tested in January 2021 compared to the sample tested in June/July 2020.

Participants
We calculated the sample size required to achieve 80% power to detect a significant Logos (MacDonald's, Kellogg's, Mastercard, Pepsi,

Apparatus and stimuli
To create and host the experiment, the online behavioral science plat- proportions. In addition, they were spaced separating the two halves on the horizontal axis (occupied area by the spaced logos: min 57120 pixel, max 61740 pixel). Each of the 12 logos (6 original and 6 spaced, see

Procedure
Each trial began with the presentation of a black fixation cross in the center of a white screen (1000 ms), followed by a central stimulus that remained visible until a response was provided (or for a maximum of 2000 ms). Subsequently, a blank was presented (500 ms). Participants were instructed to maintain the fixation at the center of the screen and to respond according to the size of the logos by using their left and right index fingers. They were instructed to press a left key if the logo was small and a right key if the logo was large (respectively the "e" and "o" keys of a QWERTY keyboard without the numeric keypad; and the "y" and "p" keys of a QWERTY keyboard with the numeric keypad). 2 Together with instruction, they were given a sample logo (different from the experimental one) with a small and large size to understand the task. Instruction emphasized both speed and accuracy of responses.
Each stimulus was repeated 10 times, thus a total of 240 trials were presented pseudo-randomly (i.e., random order controlled within participants) across two blocks of 120 trials each. A short rest was allowed between blocks. The experimental section was preceded by a practice session of 12 trials.
At the end of the experimental phase a list with the names of 18 brands (6 from the study plus 12 additional logo names) was presented to verify the effective recognition of the logo and its associated brand, with the purpose of estimating the familiarity of the stimuli. All the additional logos belonged to the same categories as the ones used in the study (i.e., letters, symbols, social images). Participants were required to identify at least 4 out of 6 brands presented in the experimental phase to be included in the analysis.

RESULTS
Response times (RTs) that were 2 SD faster or slower than each participant's mean (11.9% and 13.7% of the total trials for the 2020 sample and the 2021 sample, respectively) and errors (4.5% and 4.8% of the total trials for the 2020 sample and the 2021 sample, respectively) were excluded from the analysis on RTs. The data are available at: https: //osf.io/5p4au/?view_only = 889d10335c404790aa07a2ff5ce29d99.
Seven participants were excluded from the 2020 sample: Four reported a total number of discarded trials (including faster or slower RT and errors) that exceeded the threshold of 1 SD from the overall mean of discarded trials (29.23%) 3 ; two participants identified less than four of the six brands (66.6%) presented in the experimental phase; one participant did not complete the experiment within a single session. Therefore, the analysis was conducted on 28 participants Error bars indicate standard errors of the mean adjusted for within-participants designs (Loftus & Masson, 1994). The magnitude of the distancing effect, for each Logo, is reported on top of bars. Asterisks denote significant effects (*p < .005) social distancing act on a phenomenon deeply rooted in human behavior, that is, social interaction (Levinson & Enfield, 2006) and we may assume that the more a phenomenon is rooted in our behavior, the less people may be willing to accept a long-lasting change concerning it. This would explain why participants from the 2021 sample showed a distancing effect whereas participants from the 2020 sample did not. To conclude, the present study suggests that the acquired distancing behaviors imposed by the pandemic spread affected our performance on brand logos processing, especially when, through spaced social logos, distance in social interaction is evoked. This is the dimension mostly affected by the restrictive measures applied during the pandemic, and being social interaction deeply rooted in human behavior (Ciardo et al., 2015;Levinson & Enfield, 2006), a stronger effect emerges in participants who have been exposed longer to restrictive measures of social distancing.
These findings highlight the novelty of our study, demonstrating how social distancing can have potential repercussions on the perceptual processing of stimuli in the environment that surrounds us.
Importantly, the present study shows the role of the wide pandemic contest on perception, two phenomena that apparently are not related to suppose any influence of the first on the second, but that are somehow connected.
To make our provisional conclusion stronger, further studies will proceed with a stronger validation of the logos classification and with a larger number of stimuli.

FUNDING
This research was not supported by any funding.

DATA AVAILABILITY STATEMENT
The data are available at: The data are available at: https://osf.io/ 5p4au/?view_only=889d10335c404790aa07a2ff5ce29d99

PEER REVIEW
The peer review history for this article is available at https://publons. Several studies indeed demonstrated that information related to object size is automatically spatially coded: small stimuli are represented on the left and large stimuli on the right (see for example Walsh, 2003).