Understanding the Language of Internet Memes

Over the past week, I have fallen in love with Patti LaBelle–or her pies, to be exact. Let me explain. Last week, as I scrolled through my feed on Facebook, I found this video review of a sweet potato pie that singer Patti LaBelle is currently selling in US Walmart stores. With millions of views and a TV news interview to his credit, it is clear that this video and its creator James Wright are viral sensations. Many people created video responses to Wright’s review. Some also created image macro memes (like this one)—another common form of content shared on various social media platforms. An image macro refers to the rules for adding text to an image (Davison, 2012). But how do we understand the meaning of the sentences that image macro memes carry? How can this unique form of communication help us understand our ability to understand other sentences? Past language research suggests at least one way to answer these questions.

Some scientists believe we understand sentences by first making sense of their structure. Others believe we simultaneously analyze sentences for their meaning. Goldstein (2011) defines these ideas, respectively, as the syntax-first approach and the interactionist approach. Tanenhaus and coworkers (Goldstein, 2011) conducted an experiment that, like memes, relied on the juxtaposition of images and sentences. Participants were presented with a set of objects as they listened to recorded instructions related to those objects. Although all the participants listened to the same instructions, some were presented with a slightly different set of objects. To measure their comprehension, Tanenhaus tracked their eye movements. The syntax-first approach predicts that their eye movements of participants in both conditions would be the same since the structure of the sentences is identical. The eye movements across conditions were significantly different, however. This result implies that our understanding of sentences is not determined by syntax alone, but is also influenced by other information, like the scene we happen to be observing.

An image macro meme carries cultural information. Formally, it can be defined as “a piece of culture, typically a joke, which gains influence though online transmission (Davison, 2012).” In other words, image macro memes communicate humor based on shared cultural knowledge. The meme linked to above assumes the viewer knows who Aretha Franklin is and why her purportedly tense relationship with Patti LaBelle would be relevant to LaBelle’s pies. Since knowledge of LaBelle’s pies and her feud with Franklin be considered esoteric, I will offer a simpler example: Advice Dog. The rules of this meme require a would-be jokester to 1) start with a specific image of a dog in the center of a rainbow, 2) write one line of advice at the top of the image, and 3) write a second line of advice (usually a punch line) at the bottom of them image (Davison, 2012). Like the recorded instructions used in Tanenhaus’s experiment, this meme communicates a consistent message—funny advice. Additionally, Advice Dog provides visual stimuli which might influence comprehension the same way the sets of objects did for Tanenhaus’s participants.

So, what could this image macro meme experiment look like? Perhaps, it could be set up as a timed task. Participants would be primed with an Advice Dog meme that follows the abovementioned rules. Then, after a short delay, they would be presented a meme containing the same image as the prime stimulus, this time imposed with a sentence that follows the rules (related), doesn’t follow the rules (unrelated), or is ungrammatical (non-sentence). The participants would complete several trials, indicating whether the meme contained a grammatical sentence. The syntax-first approach would predict similar reaction times for both grammatical sentences. However, I would be curious to see if the reaction time for the related condition was significantly shorter than for the unrelated condition. If it was shorter, I think it would provide further evidence that context—a visual stimulus, in this case—is processed along with syntactic information. Maybe then knowledge of Patti’s pies would move from the esoteric to the mainstream and scientists, as well as psychology students, could have a reason (or a second one) to appreciate this legend.


Davison, P. (2012). The language of internet memes (M. Mandiberg, Ed.). In The social media reader (pp. 120-134). New York: New York University Press.

Goldstein, E. B. (2011). Language. In Cognitive psychology: Connecting mind, research, and everyday experience (pp. 306-308). Australia: Wadsworth Cengage Learning.

Leave a Reply