The vampire effect of smartphone swiping: how atypical motor actions increase ad attention but impair brand recall

Abstract

Consumers’ swiping behavior largely determines their exposure to social media advertisements. According to embodied cognition and enactment theory, advertisers might leverage atypical swiping to increase attention and thus brand recall. To identify typical smartphone swiping, the authors develop a taxonomy of the motor actions consumers exhibit when browsing social media in real life. A mobile eye-tracking experiment then reveals how the typicality of smartphone swiping affects participants’ advertising reception. The results indicate that atypical smartphone swiping increases consumers’ visual ad attention but, surprisingly, decreases brand recall. These findings, observed under realistic viewing conditions, suggest a motoric vampire effect of atypical swiping: It appears to demand the allocation of cognitive resources to the odd motor action, which diverts cognitive resources away from the ad. Thus, atypical swiping poses a threat to advertising effectiveness, and advertisers need continued research to identify ways to mitigate these negative effects.

Keywords:

• human–smartphone interaction
• social media
• embodied cognition
• motor actions
• mobile eye-tracking
• marketing effectiveness

Introduction

In recent decades, the Internet has transformed from a computer-mediated environment (Yadav and Pavlou 2014) to an environment facilitated by mobile devices (Wolf 2023). In the same period, the smartphone has become the most important advertising device available to marketers (DataReportal 2023; Ericsson 2021). Human–smartphone interactions largely rely on hand and finger movements, which in turn determine consumers’ experience of social media as they move the displayed content in and out of the smartphone screen. Yet in the extensive research devoted to social media advertising (e.g. Boerman, Willemsen, and van der Aa 2017; Gavilanes, Flatten, and Brettel 2018; Hollebeek, Glynn, and Brodie 2014; Knoll 2016; Voorveld 2019), we find little consideration of how people actually use their hands and fingers to hold and interact with their smartphones when browsing through their social media feeds and how these motions might affect attention to social media advertising.

Attention is a critical prerequisite of any advertising processing and its downstream brand effects, such as message conveyance or ad and brand recall (Pieters, Warlop, and Wedel 2002; Rossiter, Percy, and Bergkvist 2018). When consumers encounter advertising on social media through their smartphones, their attention tends to be exceptionally limited; for example, they view advertisements on Instagram for only about 1.8 seconds on average (Borgmann, Kopka, and Langner 2022). This attention toward social media ads also depends on their motor actions. Advertisements appear and disappear from the screen only when consumers use their fingers to swipe the display and navigate the feed. That is, ad exposure strongly depends on the speed of consumers’ swiping. If they swipe fast enough, consumers might even scroll past ads without noticing them. In addition to traditional attention-getting tactics based on features such as the ad’s size, pictorial elements, or color (e.g. Dukes and Liu 2023; Fernandez and Rosen 2000; Lohse 1997), we propose that user interactions, in the form of smartphone swiping, can determine consumers’ attention to social media ads.

To increase ad attention, advertisers might distract consumers from habitual fast swiping by requiring atypical smartphone swiping interactions. Such atypicality can be induced, for example, by new ad formats (e.g. carousel ads) or new social media platform features that force consumers to engage in atypical swiping. Automatic movements and tasks, as result from typical smartphone interactions, take place faster (Fontani et al. 2007) and likely result in faster swiping speeds. Because atypical smartphone swiping involves motor actions that are not automated, they should result in slower swiping speeds and allow for more attention toward ads. Considering extant advertising research that indicates a positive effect of attention on recall (e.g. Boerman, van Reijmersdal, and Neijens 2015; Pieters, Warlop, and Wedel 2002), we posit that slowing down smartphone swiping also might increase ad and brand recall.

With this research, we seek to test this prediction by exploring the role of atypical smartphone swiping on consumers’ advertising responses, and specifically whether atypical smartphone swiping enhances attention and brand recall. In turn, we make two main contributions to extant literature. First, we develop a taxonomy of the combinations of the holding grip and input finger that people use to browse social media, in a pre-study conducted under realistic viewing conditions (De Pelsmacker 2021) that involve participants in their own homes, using their own devices and social media feeds. In this taxonomy, we identify both typical and atypical human–smartphone interactions that occur during people’s uses of social media. Second, we investigate whether and how atypical smartphone swiping affects consumers’ attention to and brand recall for advertised brands. Building on the pre-study results, we conduct an eye-tracking experiment, again under realistic viewing conditions. The results confirm that atypical smartphone interactions increase the total gaze duration on social media ads, but in contrast with our prediction, they decrease brand recall. In this sense, we provide the first evidence of a motoric vampire effect of atypical smartphone swiping, which should inform both advertising and psychological motor action research, as well as the practices and designs adopted by advertisers in social media settings.

Theoretical background and hypotheses development

Motoric human–smartphone interactions

The way people use their hands to interact with the world has been the topic of studies in various fields, such as touch (Krishna, Luangrath, and Peck 2024), haptics (e.g. Feix, Bullock, and Dollar 2014; Klatzky, Lederman, and Metzger 1985; Lee et al. 2016; Napier 1956), and human–computer interactions (e.g. Kim and Jo 2015; Tsai, Tseng, and Chang 2017; Wang et al. 2019). Over the course of evolution, humans developed superior fine-motor skills, enabling them to use their hands to explore and interact with objects and tools (Luangrath et al. 2022). In general, motoric interactions consist of either holding grips to grasp an object or uses of input fingers to adjust it. Napier (1956) further classifies holding grips into a power grip for stability and a precision grip for sensitivity or accuracy. Lederman and Klatzky (1987) also identify typical hand movements that people use consistently to explore everyday objects. Over time, they develop motoric preferences for such interactions, leading to natural grip patterns when using handheld tools such as a pen, a coffee mug, or a phone (Norman 1988).

As human–smartphone interaction studies reveal, the index finger and thumb are the main choices exhibited by consumers (Shin et al. 2016), though they can use other fingers as well. Whether people use the fingers on their right or left hand generally is predetermined by their dominant hand (Miyaki and Rekimoto 2009), that is, the hand they use for writing. Although early model mobile phones varied significantly in design (e.g. clamshell phones, Blackberries with full keyboards), the iPhone has defined the general design of a modern smartphone: large touchscreen, reduced buttons, and size and weight dimensions that allow for single-handed navigation. How people interact with these largely standardized modern smartphones also has attracted research in the human–computer interaction domain (e.g. Kim et al. 2006; Perry and Hourcade 2008; Wang et al. 2019). Laboratory experiments have sought to understand how users’ age (Tsai, Tseng, and Chang 2017), choice of input finger (Kim and Jo 2015), grip variations (Lee et al. 2016), or handedness (Chen, Zhu, and Yang 2023) can affect their gesture operations. Although these studies offer some insights, we know of no comprehensive overview of typical motoric smartphone interactions as they occur during real-life social media usage. Therefore, to understand typical versus atypical smartphone swiping, we conducted a pre-study in which we explore which grip and finger choices consumers make and prefer when browsing their social media newsfeeds on their smartphones. Building on these results, we then induce atypical smartphone swiping in an eye-tracking experiment, such that we can analyze whether atypical swiping influences people’s attention and brand recall.

Visual attention and embodied cognition theory

For an ad to work, it needs attention, but consumers’ attention is exceptionally limited in cluttered social media environments (Bergkvist and Langner 2023; Beuckels et al. 2021; Nelson-Field, Riebe, and Sharp 2013). Not only are attention spans decreasing, but more social media ads constantly join the battle for attention (Duff and Segijn 2019). While people are using their hands to swipe through their social media newsfeeds on their smartphones, they also are exposed to advertisements, primarily through their eyes, making visual attention an essential element (Van Raaij 1989) for assessing whether and how long ad exposure actually occurs (Frade, de Oliveira, and Giraldi 2023).

Embodied cognition theory also asserts that attention is connected to bodily experiences and motor actions (Rizzolatti et al. 1987), and mental processes are influenced by physical experiences. That is, people direct their visual attention toward the location of their planned hand movement (Abrams et al. 2008). In turn, we argue that attention to social media ads on smartphones is a mental process that also is affected by goal-oriented hand movements (Festman et al. 2013; Goldinger et al. 2016). Exposure times for social media ads depend on consumers’ individual motoric interactions, but with extensive practice, motor commands and muscular activity patterns form that allow for skilled, fluent performance (Land et al. 2013). When movement control in skilled motor tasks becomes automatic, motor actions are usually performed faster (Fontani et al. 2007), so swiping on smartphones makes ads appear and disappear at different rates, with strong impacts on visual attention. Atypical swiping, which is not conducted automatically, occurs more slowly and consciously, which in turn should increase people’s exposure to social media ads. Thus, we hypothesize:

H1: Atypical swiping leads to (a) longer total gaze durations and (b) more fixations on social media ads than typical swiping.

Duration of exposure and brand recall

Measures of advertising effectiveness often rely on brand recall and attitude formation (e.g. Van Raaij 1989). Brand recall pertains to consumers’ brand awareness, such that it represents their ability to remember an advertised brand (Rossiter, Percy, and Bergkvist 2018). In turn, it requires sufficient time and opportunity for the consumers to cognitively process the advertisement (MacInnis and Jaworski 1989; Wilson and Till 2012; Wilson, Baack, and Till 2015). Longer advertising exposures increase the learning of the ad and brand (Pieters, Warlop, and Wedel 2002; Singh and Cole 1993; Newstead and Romaniuk 2010). Attention in turn is essential, such that a lack of attention is the main hindrance to advertising effectiveness (Liu-Thompkins 2019), whereas greater attention generally results in higher brand recall (Bergkvist and Langner 2023; Guitart, Hervet, and Hildebrand 2019; Rossiter and Percy 2017; Simmonds et al. 2020). Thus, longer exposures to an ad, due to atypical swiping (H1), should result in better brand recall.

Memory of motor actions and processing depth

Attention and brand recall also might be influenced by the motor actions (e.g. swiping) that consumers perform during an ad exposure. Macedonia et al. (2019) demonstrate that even if they just observe motor actions, people’s processing depth increases, as revealed by fMRI scans that showed the recruitment of new cortical areas. Motor-action and learning theory also predicts an enactment effect, such that memory of self-performed motor actions is superior to other types of learning (Engelkamp and Zimmer 1994). Similarly, performing motor actions during encoding processes enhances people’s memory of words, as Macedonia and Knösche (2011) establish by inducing students to make gestures while learning words in a foreign language. Evidence of drawing effects similarly indicates that drawing pictures during encoding boosts subsequent memory (Fernandes, Wammes, and Meade 2018; MacLeod and Bodner 2017).

Accordingly, we anticipate that information perceived during the performance of novel motor actions (e.g. atypical smartphone swiping) may benefit from the higher-level processing induced by the execution of these motor actions. If atypical swiping enhances the level of processing, it also might stimulate learning of social media ads encountered during atypical swiping sessions. In detail, we predict that atypical swiping increases both exposure time and the level of processing, which together stimulate greater learning (recall) of brand names advertised in social media ads.

H2: Atypical swiping leads to higher brand recall than typical swiping.

Pre-study: identifying typical and atypical smartphone swiping

With the pre-study, we seek to identify typical and atypical smartphone interactions people have when using social media in a real-world setting. Building on research into haptics and human–smartphone interactions, we specifically investigate dominant grip and finger choices during social media smartphone interactions. In line with Norman (1988), we expect that consumers have developed a motoric preference for a specific dominant grip and finger choice. Therefore, we seek to derive typical and atypical smartphone interactions for individual consumers.

Method

Participants

Thirty participants, aged 19 and 69 years, took part in the study (43% women, M_age = 35.4 years, SD_age = 14.1). Most participants were employed (60%), and the rest (40%) were university students. The participants received a 20 EUR Amazon gift voucher for their participation.

Study design

The pre-study combines in-home videography to observe participants’ hand movements with qualitative interviews, in which they verbally described their smartphone usage while swiping through their social media feeds.

Procedure

To start, we asked participants to use their smartphones as they normally would to browse social media. We required that they use their own smartphones to browse their own social media newsfeeds (e.g. Instagram, Facebook). These individual sessions were not restricted in time and lasted between 1:22 and 6:32 min (M = 3:51 min). We then engaged them in a think-aloud interview, in which they commented on their own video-recorded swiping behavior while watching the video. Following the interview, we posed open-ended questions such as “Please describe how you hold your smartphone and swipe or scroll when using [social media platform],” and “When you use the same social media platform in different situations, e.g. during breakfast, in public transportation, while watching TV, do you hold and swipe differently?”

Data analysis

In total, we recorded 2 hours of typical smartphone usage, along with more than 14 hours of follow-up qualitative interviews. In the first step, one author analyzed the video material to identify how participants used their hands to hold the smartphone and their fingers to swipe through their social media feed. In turn, we could define interaction episodes comprised of holding grip and input finger combinations (e.g. holding the smartphone in the right hand and swiping the newsfeed with the thumb of the same hand; Figure 1). In the second step, we analyzed the interviews to identify how the participants experienced typical and atypical uses of their smartphone. Both steps were supported by MAXQDA.

Figure 1. Taxonomy of motoric human–smartphone interactions.

Results

Taxonomy of motoric human–smartphone interactions

Participants used different holding grip and input finger combinations to swipe through their social media newsfeeds on their smartphones. In total, we observed 94 interaction episodes of holding grip and input finger combinations, which we illustrate in Figure 1, according to a proposed taxonomy.

Holding grip

Participants held their smartphone either with one hand (n = 51 of 94 interaction episodes) or two hands (n = 43). In the case of one-hand grips, they clearly preferred the right hand (n = 37) over the left hand (n = 14). Participants who held the device with two hands exhibited either a dominant hand (n = 34) or equivalent holding between both hands (n = 9). If they used a dominant hand, in most cases, they held their smartphone in one hand and used the additional hand for support. Similar to the one-hand grip, the right hand was dominant more often (n = 31) in the two-hand grip than the left hand (n = 3). Holding the phone equally with both hands only occurred during texting, not when participants swiped through their newsfeeds. Rather, while swiping newsfeeds, they used either one hand or two hands with a dominant hand.

Input finger

Participants used one or two fingers to interact with their smartphones. One-finger input (n = 84) was far more common than two-finger input (n = 10). In most cases, the thumb (n = 69) was the primary choice for swiping, and the right thumb (n = 64) was used more often than the left thumb (n = 5). The second most commonly used finger was the index finger (n = 14), again primarily right (n = 11) instead of left (n = 3). One participant used the right middle finger for inputs, but no other one finger inputs occurred in our observation. Participants engaged in two-finger input (n = 9) used both thumbs at the same time for texting, while holding the phone with both hands. Only one participant jointly used the right index and middle finger at the same time for swiping. No other two-finger input combinations occurred in our observations.

Typical and atypical smartphone swiping in social media

With these observations, we identify two basic types of smartphone interaction when swiping through newsfeeds: thumb-swiping while holding the phone in the same (dominant) hand or index-swiping using the dominant hand while holding the phone in the non-dominant hand. The clear preference for the right over the left hand for holding the smartphone and for providing input to the device reflects participants’ handedness. The various observed interaction types represent modifications of these two basic types. For example, for participants who used two hands, the additional hand functioned as support; the basic interaction type continued to entail holding the phone in the dominant hand and using the thumb for input, as explained by one participant (Table 1):

Table 1. Think-aloud statements related to typical and atypical motoric human–smartphone interactions.

Participants	Statement related to Typical and Atypical Smartphone Interaction	Human–Smartphone Interactions
P 02, female, age 19	I am left-handed. I always hold it in my left hand and swipe and write with my [left] thumb.	Holding grip, one hand, left hand
P 15, female, age 38	I do everything [on the phone] with one hand only. I actually hold it all the time like this. I thought again today of my poor right hand. So, I actually tried it briefly in my left hand, but immediately I switched back to my right hand. I’m just through and through right-handed there.	Holding grip, one hand, right hand
P 03, male, age 36	Yes, I am left-handed. I hold it in my left hand. For the moment, I also hold it with my right hand for support. I would say it’s heavier, but larger. I actually hold it with two hands now.	Holding grip, two hands, dominant left hand holding
P 30, female, age 22	This is now my fourth generation iPhone. It only used to be like that with the narrower ones [holding it only in the right hand]. Since I have the larger one, because I just have small hands, I have to support with my left hand as well. And I also notice after a certain time it gets heavy.	Holding grip, two hands, dominant right hand holding
P 16, female, age 19	When I write, I always take the other hand to it and then I sort of hold it like that with both hands. Also, if I were to comment, then I would write like this [with both thumbs].	Holding grip, two hands, equally holding
P 03, male, age 36	The left thumb, yes. The left thumb is the thumb to control.	Input finger, one finger, left thumb
P 13, male, age 24	I always keep it this way. I never use the apps in landscape mode from the beginning. Always with the right hand. Even in bed, always in portrait mode with the right hand. And always with the right thumb.	Input finger, one finger, right thumb
P 22, female, age 29	Exactly, and also through the newsfeed. I actually scroll down most of the time. Now that wasn’t unnatural, it’s actually always like that, always with the [left] index finger no matter where I am.	Input finger, one finger, left index finger
P 09, male, age 41	Yes, that’s how I prefer to do it. I hold it in my left hand and use my right finger [for input]. Because one-handed operation is always a bit uncomfortable with today’s smartphone sizes. And that’s why I try to use both hands whenever possible.	Input finger, one finger, right index finger
P 19 male, age 43	Yes, with the [right] middle finger. I do that everywhere, too. It’s very, very typical. I think also because the middle finger is perhaps my longest finger. That’s the most comfortable way. That’s actually the most comfortable position.	Input finger, one finger, right middle finger
P 29, female, age 61	So, truth be told, it looks like my children told me I had to learn to text with both thumbs. Because I like my keyboard very fast, so they’re like, no, you’ll have to use two thumbs.	Input finger, two fingers, both thumbs, equally inputting
P 26, female, age 69	I usually do this with my middle finger [and index finger]. Like this from bottom to top.	Input finger, two fingers, right middle finger

All interviews were translated into English using a back-translation approach.

When I’m tired, I sometimes also use my left hand for support, but I still hold my phone in the right hand and swipe with my right thumb. (Participant 8, male, age 38)

The use of the right middle finger or the combination of the right middle and index fingers similarly can be categorized as holding the phone in the non-dominant hand and using a primary finger for inputs. Only in very rare cases (n = 2) did participants use their middle finger; most participants used their index fingers. Notably, some participants selected smartphone accessories to match their typical smartphone interaction, such that several thumb swipers installed so-called pop-sockets that support single-handed holding of the phone, and some index swipers chose a foldable phone wallet that helps them hold the phone in one hand and use the index finger for inputs. A participant explained the use of his smartphone accessory by noting:

I do have an extra case for my phone. My wallet case might be special. I swing it open and use it like this [holding it in the left hand and using the index finger for inputs]. I also call with the phone like this, and it holds all my credit cards. I just need to grab my phone [and case] and can go shopping and everything. I find this very convenient. (Participant 19, male, age 43)

When switching to texting movements, such as to write a comment, participants either kept the same interaction type and used one thumb or the index finger or switched to two-thumb typing. However, once they returned to browsing the newsfeed, participants switched back to their typical swiping types. Preference for either of the two basic types seems to be robust, such that one swiping type usually represents typical swiping and the other represents atypical swiping for a particular individual. None of the participants changed between the two types during our study. Six participants indicated index swiping and 24 participants thumb swiping as their typical smartphone swiping when browsing social media. All expressed a clear preference, as exemplified in the following statements:

This is a very typical grip. My left hand is holding the phone. That’s very typical for me, and I use the index finger of my right hand to do something on the phone. I use my [right] index finger also for texting. I really only work with one index finger. (Participant 20, male, age 58)

I always hold my smartphone in my right hand just like this. I hold it relatively straight up and I am a thumb-swiper. Also, when I am walking outside or riding the bus. I always have it in one hand and swipe with my thumb. (Participant 27 male, age 40)

In addition, one participant reported that his swiping does not feel familiar yet, because he just switched to a different phone model:

For two weeks I have a new phone now. This is a new phone. I wouldn’t say it’s much heavier, but a bit larger in size. Holding it and swiping still feels a bit unfamiliar. (Participant 3, male, age 36)

Discussion

With a taxonomy of all grip–finger combinations, we can identify dominant motoric human–smartphone interactions that consumers typically use to browse their social media newsfeeds. By observing how consumers browse social media in real-life, we note their reliance on either thumb-swiping while holding the device in the same (dominant) hand or index-swiping, which involves using the dominant hand while holding the device in the non-dominant hand. All other interaction types represent modified versions of these two basic types. Moreover, consumers have developed a clear motoric preference for either one of these two basic interaction types. They learn how to navigate their smartphones and use one of the two basic types by default. Consumers stick with this preferred, typical swiping type and only seldom change to another, atypical swiping type (e.g. when tired). Similar patterns have been observed for handheld tools (Norman 1988), for which people develop a preferred usage style over time while learning how to use them.

Experiment: an eye-tracking study to analyze the impact of atypical swiping on attention and brand recall

To analyze the impact of atypical swiping on attention and brand recall, we conducted an eye-tracking experiment with participants using their own devices and swiping their own social media newsfeeds in their living rooms. Informed by the pre-study, we required them to employ thumb or index swiping, which correspond with either typical or atypical swiping.

Method

Participants

Thirty-six participants, aged 21–58 years, took part (61% women, M_age = 30.5 years, SD_age = 9.4). Most participants (81%) were employed, and the remaining 19% were university students.

Study design

With a within-subject design, we induced atypical smartphone swiping by directing participants to employ both basic smartphone swiping types (thumb-swiping while holding the device in the same hand and index-swiping using the dominant hand while holding the device in the non-dominant hand) in two distinct swiping sessions.

Procedure

The experiment was conducted between October and December 2022 in participants’ homes, using a mobile eye-tracking device (Tobii Pro Glasses 2). We instructed the participants to browse their own social media newsfeeds (e.g. Instagram, Facebook) using their own smartphones as they normally would (Figure 2). The sessions lasted approximately 2 min each and were separated by a brief interview. We randomly assigned participants to two experimental conditions, in which they were required to use their thumb (index finger) in the first and their index finger (thumb) in the second session.

Figure 2. Experimental setup in participants’ homes (left: thumb swiping; right: index swiping).

Measures

In the interviews that followed the two swiping sessions, participants rated the sessions, relative to their natural swiping behavior, on a scale from “not typical at all” [0] to “very typical” [+6]. A manipulation check revealed that in 68% of the cases, the difference in typicality scores was 4 or higher, indicating a successful manipulation and substantial differences in typicality between index and thumb swiping. For the unaided brand recall measure, we asked participants to state any brand names they recalled from the two swiping sessions, then compared those listed names with the recorded ads; verified brand names represented valid recalls. The recorded eye-tracking data were exported into iMotions software (version 9.3). In addition, we created individual areas of interest (AOI) for all sponsored posts, which we coded frame-by-frame to reflect the dynamic situation. A sponsored post was logged as viewable at the moment at least 50% of the stimulus entered the screen until 50% of it left the screen (Trabulsi et al. 2021). Attention was measured by total gaze duration in milliseconds and fixations within each AOI. The minimum duration for a fixation was set at 80 milliseconds (Boerman and Müller 2022).

Results

Impact of typicality of swiping styles on attention toward the ad

To understand how the typicality of smartphone swiping styles influences attention, we analyzed all cases of ad viewing while swiping through the newsfeed (n = 411). We winsorized one gaze duration value that exceeded 3 SDs from the grand mean by the closest inlier (Meyvis and van Osselaer 2017; Bellman et al. 2019). To estimate the effect of the typicality of consumers’ hand movements on their gaze duration and fixation frequency, we employed generalized linear mixed models (GLMM). Gaze duration was modelled to follow a Gamma distribution with a log-link (Rosbergen, Pieters, and Wedel 1997). For fixation frequency, a count variable, we used a Poisson distribution with a log-link (Pieters, Warlop, and Wedel 2002). By including the session index as a control variable, we can control for potential primacy or recency effects, according to the order of the two sessions. In addition to fixed effects, we included random intercepts to account for variation in the outcome variable that might be attributable to the participant (Barr 2013; Judd, Westfall, and Kenny 2012; Quené and van den Bergh 2008).

We find significant negative effects of typicality on gaze duration (b = −0.070, p < .001) and fixation frequency (b = −0.084, p < .001), which indicate that the typicality of hand movements impairs attention toward the advertisement, in support of H1. Figure 3 shows the estimated marginal means of gaze duration and fixation frequency for low, medium, and high values of typicality. The planned contrasts of the estimated marginal means also reveal significant differences across all three contrasts (p < .001).

Figure 3. Effect of typicality on gaze duration and number of fixations.

Impact of typicality of swiping styles on brand recall

In generalized linear mixed effects logistic regressions, we controlled for session index and random effects related to the participants. Overall, 19 of 411 brands (4.6%) were recalled in the unaided brand recall test. The GLMM indicated a marginally significant positive effect (b = 0.966, p = .055), such that typicality increased brand recall. Because we expected greater attention, due to a longer total gaze duration evoked by atypical swiping, we hypothesized a negative effect of typicality on brand recall, but instead, the planned contrasts of the estimated marginal means reveal only a marginally significant difference between high and low levels of typicality (M_low = 0.001, M_high = 0.042, p = .086). Figure 4 illustrates the estimated marginal means of brand recall for low medium and high values, which lead us to reject H2.

Figure 4. Effect of typicality on brand recall.

Discussion

This experiment analyzed the effects of atypical swiping on visual ad attention and brand recall. When participants had to use a motoric smartphone interaction that they typically would not use, they swiped more slowly and spent more time looking at social media ads (longer gaze durations and higher numbers of fixations). These results confirm our predictions and align with previous observations showing that atypical motor actions are executed more slowly and less automatically (Fontani et al. 2007; Perry and Hourcade 2008).

We further hypothesized that performing atypical swiping actions would result in improved brand recall due to increased attention and processing depth (Pieters and Wedel 2004; Singh and Cole 1993; Newstead and Romaniuk 2010). However, we find that atypical smartphone interactions impair recall of the advertised brand, despite the longer ad exposures. These findings contradict psychology research that identifies enhanced learning of words encoded while people either observe a new motor action (Macedonia et al. 2019) or perform it themselves (Macedonia and Knösche 2011). Notably, the motor actions that Macedonia et al. (2019) and Macedonia and Knösche (2011) required of their participants related semantically to the meaning of the words they were learning. The atypical motor actions performed by participants in our experiment, however, were not semantically related to the advertised brands. Thus, we posit that a moderator of the effects of atypical swiping might be the semantic relatedness of the motor actions. That is, perhaps positive effects hinge on a semantic link between the motor actions and the information to be memorized (Kormi-Nouri 1995; Li et al. 2022).

Considering the limitations of humans’ cognitive capacities (Bryant and Comisky 1978; Lavie et al. 2004), we further posit that atypical smartphone swiping requires cognitive resources, devoted to executing the unfamiliar motor actions, such that consumers are left without enough resources remaining to process the advertisement and memorize the advertised brand. It seems that processing atypical swiping requires more cognitive capacities than typical swiping, and therefore, consumers have insufficient capacities to process and memorize brand-related information available in the ads. These findings suggest a new type of vampire effect: the motoric vampire effect, resulting from atypical motor actions executed to interact with the advertising medium itself. Vampire effects are well-known advertising consequences that can arise from using celebrity endorsers (Chan and Chau 2023; Erfgen, Zenker, and Sattler 2015; Evans 1988), influencers (Waltenrath, Brenner, and Hinz 2022), or humor (Eisend 2011)—other features that increase attention to an ad but potentially impair brand recall. In the joint processing of atypical smartphone interactions and brand advertising, performing motor actions seems to hinder the encoding of brand names, such that social media ads encountered during atypical swiping situations get processed only superficially, leading to lower brand recall.

This study accordingly extends prior research on media context (e.g. De Pelsmacker, Geuens, and Anckaert 2002; Yoon, Huang, and Kim 2023), in that it emphasizes the pivotal role of motoric interactions as a media context factor in the domain of social media advertising. Furthermore, the investigation contributes to the wider field of haptics, building on foundational works by Lederman and Klatzky (1987) and Norman (1988) to broaden the concept of typical motoric interactions to apply to modern smartphones. By delineating the impact of atypical swiping on advertising effectiveness, this study enhances our comprehension of consumer behavior in social media advertising settings while also shedding new light on the complex interplay of physical interaction modalities with cognitive processing mechanisms.

Conclusion

Despite substantial interest to understanding social media advertising, questions surrounding how people actually use their hands and fingers to browse the newsfeeds on their smartphones and how these hand and finger movements affect their responses to social media advertising have not been considered. We present the first investigation of consumers’ motoric smartphone interactions and its impact on ad attention and brand recall in real-world settings, using a mixed methods approach of in-home videography and an eye-tracking experiment. We find that consumers’ visual attention increases when the motor actions they execute to browse their social media newsfeeds are atypical. However, and contrary to our expectations and predictions that longer attention times translate into higher brand recall (Pieters and Wedel 2004), we find no evidence of a positive effect of atypical swiping on brand recall. Instead, despite their faster swiping and shorter gaze durations, consumers recall advertisements better when engaged in typical smartphone swiping. Thus, brands do not benefit from the increased attention that results from atypical swiping. On the contrary, our results suggest that atypical smartphone swiping triggers a motoric vampire effect, such that consumers devote additional processing capacities to action-relevant information rather than ad content. In this sense, atypical swiping can pose a threat to advertising effectiveness.

Implications for practice

We anticipated that advertisers might induce atypical swiping by consumers, as a means to increase ad attention and thereby brand recall, but instead, a motoric vampire effect undermines the latter outcome. On the basis of our multimethod results, we offer several alternative implications for practitioners. First, inducing atypical motor actions may threaten advertising effectiveness. Adopting novel and still unfamiliar ad formats (e.g. when Instagram introduced carousel ads) might increase visual attention for ads, but it is unlikely to enhance brand memory. Advertisers therefore should be cautious about spending their media budgets in newly introduced new ad formats that require atypical smartphone swiping, at least until the gesture becomes typical to consumers.

Second, advertisers and social media platforms need to be aware that major platform changes (e.g. when Instagram introduced stories) may have strong impacts on how consumers not only navigate, but also process information. New platform features or changes will likely draw people’s attention, which in turn can have negative effects on brand advertising.

Third, we investigate currently standard smartphone designs, as established mainly through the evolution of Apple’s iPhone. New technology breakthroughs that allow for novel smartphone design (e.g. foldable screens) will likely affect existing motoric interactions or introduce new ones. Advertisers in turn should anticipate that any such design changes may alter consumers’ motoric interactions and thus their advertising effectiveness.

Fourth, our study focuses on advertising in social media, but we anticipate that the findings might apply to other, related contexts, such as mobile shopping (e.g. new ad formats in Amazon 2022), in-game advertising, advergames (e.g. Goh and Ping 2014), and mobile applications in general. In these settings too, inducing atypical motoric interactions, such as new movements or unfamiliar navigation, may overshadow content and leave less processing capacity, unless those movements are semantically related with the focal content.

Limitations and future research

In the real-world experiment, we measured ad attention and recall for any advertised brands that appeared in the participants’ actual newsfeeds. Thus, we could not control for the advertising content or brands across conditions or test any effects of typicalness with the same ads. A fully controlled laboratory experiment using standardized newsfeed content and ads could help corroborate our findings, as well as potentially explore brand recognition as a less challenging response that is also highly relevant for recognition-based purchase decisions (e.g. fast moving consumer goods purchased in a supermarket; Rossiter, Percy, and Donavon 1991). In addition, we acknowledge that atypical smartphone swiping can become more typical through extended usage, but we did not consider such potential learning effects over time.

Therefore, we encourage further research to investigate how interface designs can induce atypical motor actions, as well as explore motoric interaction types involving other screen-based devices (e.g. tablets, VR/AR headsets). Certain advertising formats also might alter consumers’ motoric interactions with social media ads and distract them from habitual fast swiping. Furthermore, to mitigate the negative effect of atypical swiping, brands might attempt to create a semantic relation between swiping gestures and brand assets, such as by prompting consumers to trace the brand logo in a social media ad. With such an approach, brands arguably might benefit more from motoric smartphone interactions, in line with the drawing effect (Fernandes, Wammes, and Meade 2018; MacLeod and Bodner 2017; Schwartz and Plass 2014). Finally, atypical smartphone swiping occurs regularly in consumers’ everyday lives, whether due to the adoption of new devices, new ad formats, or changes in social media platform features. Therefore, we call for continued research that addresses specific applications and identifies ways to mitigate the negative effects on advertising effectiveness.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Due to its proprietary nature, the dataset used in this study cannot be made openly available.

Additional information

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sector.