*#foodie: Implications of Interacting with Social Media for Memory*
Jordan Zimmerman & Sarah Brown-Schmidt (Vanderbilt University)
jordan.e.zimmerman@vanderbilt.edu
Social media is an increasingly popular outlet for social communication. On
many social media platforms, the user experience involves commenting on or
responding to user-generated content such as images of cats, food, and
people. In 2 experiments, we examine how the act of commenting on social
media images impacts subsequent memory for those images, using Instagram
posts as a test case. This project was inspired by recent findings from
laboratory studies of conversation that found that describing a picture for
a conversational partner boosts recognition memory for those images1,2.
Here we aim to understand how this finding translates to the more
ecologically valid realm of social media interactions. A 2nd motivation for
the study is the popularity of food and dieting related content on
Instagram, and prior findings that use of Instagram in particular is
associated with disordered eating behavior3-5.
In* Experiment 1 *(N = 100) participants (Ps) were recruited thorough
Mechanical Turk. Average age was 37 years; Ps reported gender as female
(n=46), male (n=53), genderqueer (n=1). The materials were assembled by
perusing a large number of Instagram posts. Posts featuring food served as
critical stimuli, and were further divided into the categories of “healthy”
(e.g. fruit, salad) and “unhealthy” food (e.g. brownies, fries). Posts
featuring dogs, cats, nature were used as control images. In total, Ps
viewed 100 Instagram posts, 20 in each of the 5 categories. For half of the
posts, Ps were prompted to provide a comment using the same phrasing as on
the Instagram platform (e.g., “Add a comment…” à P types: “*yummy I want
one!”*). After viewing the 100 Instagram posts, Ps completed 17 math
problems as a distractor task. Ps then completed a recognition memory task,
viewing a series of 200 images, half of which were old and seen in the
exposure phase and half of which were new, and responding whether they had
seen the picture in the prior phase or not. Across experimental lists, we
counterbalanced which images were old/new, and which images were commented
on vs. not. Lastly, Ps completed the Eating Disorder Examination
Questionnaire (EDE-Q) to assess eating disorder symptomology. *Results:*
Memory data were analyzed using a generalized mixed effects model. Memory
was good (*z*=22.11, *p*<.0001); healthy food was remembered less well than
control images (z = -2.87, p < .01), unhealthy food was remembered *better*
than control (z = 2.69, p < .01). Previously viewed images were more likely
to be correctly recognized if P had commented on them (*z* = 18.09,
*p*<.0001)—for
each additional word in the comment, odds of correct recognition was 1.37
times greater (*z* = 6.09, *p*<.0001). Model based estimates of the
reliability of these effects revealed a lack of stable individual
differences in memory for food, preventing a meaningful test of the
relationship of memory for food-related posts with EDE-Q.
*Experiment 2 *(N = 150) was a direct replication of E1, except the sample
was restricted to Ps who identified as female and between the ages of
18-30, a sample with higher risk for disordered eating behaviors6-8.
*Results:* As in E1, memory was overall good, (z=30.23, p<.0001), with
unhealthy food remembered better (z=8.81,p<.0001). Commented-upon images
were remembered better (*z* = 37.41, *p*<.0001), and for each additional
word produced in the comment, the odds of correct recognition were 1.20
times greater(z=4.61, *p*<.0001). As in E1, the memory effects were low in
reliability by-participants indicating a lack of stable individual
differences in memory for Healthy and Unhealthy food in particular, thus
preventing examination of a relationship with EDE-Q.
*Discussion:* Across 2 experiments we observe that commenting on Instagram
posts consistently boosts subsequent recognition, and that correct
recognition increases with comment length. These findings extend to social
media interaction prior findings that generating and producing information
9,10 including linguistic information in conversation,1,2,11 boosts memory
for it. In both E1-E2, “unhealthy” food images such as chocolates were
particularly well remembered, however these memory findings did not relate
to self-reported eating behavior. Taken together, our findings show that
the way in which we communicate and engage with social media content shapes
subsequent memory for it, raising new questions about how our online lives
persist in memory over time, potentially shaping future behavior.
#foodie_Implications_video.mp4
<https://drive.google.com/a/vanderbilt.edu/file/d/1BmDrdGg-w4ilMCKsxwqWqCCQtiCguMPv/view?usp=drive_web>