Psychological Priming, Indoctrination, and a Crises in Replication
I n cognitive psychology, the priming effect is the effect of being influenced in how one later responds to a stimulus upon being previously exposed to another; typically, it occurs without one’s awareness (American Psychological Association, n.d.). For example, if one is exposed to the word “doctor”, one will recognize the following word “nurse” more quickly than following the word “bread”. The priming effect is important because it aids in the activation of information by making related ideas more easily accessible. While much has been discovered about priming effects since the latter half of the twentieth century, there is still much to be understood. For instance, the possible confounding of experimenters’ expectations for the priming effect makes it difficult to determine what causes the priming effect. Coupled with a lack of research replication, it may be argued that we have only scratched the surface of unraveling the mind, its stores, and how it may be influenced by others.
Priming Effects Research
There are several types of priming, such as semantic priming (word associations by linguistic reason, e.g., “dog” and “bark”), associative priming (a stimulus and a response that are often paired with one another), repetition priming (a stimulus and response that are repeatedly paired, as seen in classical conditioning), perceptual priming (pairing two words that are textually similar, e.g., “food” and “mood”), conceptual priming (two words that are often conceptually paired, e.g., “table” and “chair”), and masked priming (where priming effects are still observed despite one word being partially obscured, e.g., “insect” and “r — ”). By being exposed to the former stimulus, one will be inclined towards the latter stimulus in the above examples. So, for instance, if one is prompted to hear either “yanny” or “laurel” before listening to an aural ambiguous recording, one is more likely to respond with that word or semantically related words. Priming effects are understood to serve an important role in long-term memory by making more related information to a target accessible when activating said target (Schwartz, 2018).
Priming Effects vs. Expectation
Interestingly, research suggests that priming effects can be used to not only help activate connections between different pieces of information, but also elicit desired responses or behavior from others (Masling, 1965). However, little is known about how much the priming effect is responsible for this phenomenon and how much is driven by expectation, since research also suggests that it may be just as influential (Doyen, 2011; Masling, 1965).
Masling (1965) sought to determine whether the effects of priming were strong enough to not only influence test examiners, but the participants that they examined. Masling hypothesized that participants’ responses would conform to the primed examiners’ expectations of what constituted a desirable response, and secondly, that examiners would subconsciously provide more verbal reinforcement for those responses they wished to receive than for those they did not wish to receive. The independent variable (IV) was the type of indoctrination the examiners received, and the dependent variable (DV) was the kind of response elicited from the test-takers. The researchers measured this by priming Rorschach examiners with falsified research to expect either a human or animal type of interpretation (anything pertaining to a human or animal part, product, symbol, or affect) from the test-takers, followed by counting the number of human-to-animal responses from the test-takers.
The examiners were randomly assigned into two experimental groups: one who would be led to believe that experienced examiners always produced more human responses than animal responses from their subjects, and the other vice versa. Based on their design, they predicted that the animal indoctrination group of examiners would influence their test-takers to “see” more animal features in the inkblots, while the human indoctrination group would influence their test-takers to “see” more human features in the inkblots.
The results revealed that the examiners who were told that animal responses were more desirable than human responses obtained more animal responses than the group of examiners told the opposite (p = .04), while the other group who were told that human responses were more desirable than animal responses obtained more human responses than the latter group (p = .05). Additionally, the frequency of reinforcement of human and animal responses between the two groups of examiners was found to be statistically insignificant.
According to Masling (1965), these findings are consistent with the hypothesis that the participants’ responses would conform to the examiners’ expectations, but not the secondary hypothesis that examiners would provide more verbal reinforcement for those responses they wished to receive. It is unknown to Masling why a difference exists between the examiner groups in the total number of reinforcing comments made and the lack of statistical insignificance, but he suggested that it must be partially due to nonverbal reinforcement that could not be captured in the tape recording.
Even though there was a lack of verbal reinforcement demonstrated in Masling (1965), it is certainly possible that novel, nonverbal information can be acquired via priming (Musen & Squire, 1992). While this nonverbal information was presented in the form of textual words and geometric illustrations, other research demonstrates that priming can occur in within-modality and across-modality conditions for both amnesiacs and normal individuals (Graf et al.,1985), and that information about discrete emotions are shared across nonverbal modalities (Pell, 2005).
Recent research also suggests that despite Masling (1965) finding of an association between the type of priming experienced by examiners and the responses elicited from students, the examiners’ expectations for the kinds of responses they wanted may have influenced the students just as much as the priming they received.
Bargh et al. (1996) sought to examine the automaticity of attitudes and social behavior by studying the effects of trait construction and stereotype activation on one’s actions. They hypothesized that social behavior is subconscious and is therefore capable of automatic activation by the presence of environmental stimuli. Consistent with their hypothesis, the results of their first two experiments revealed that the mere presence of words associated with personality traits were enough to automatically prime the study’s participants to portray those traits in their behavior. However, the results of their third experiment were statistically insignificant for the activation of the stereotype of a Black man based on the participants’ level of prejudice (Bargh et al., 1996). Doyen et al. (2011) questioned whether or not the examiner’s expectations may have also influenced the participant’s behavior, as well as the assumption that high-level semantic priming can occur both automatically and subconsciously when research suggests otherwise. Because of this, they replicated the method of Bargh et al. (1996) but found results to be inconsistent with the previous study. Where no priming effects were observed from the first experiment, the second experiment demonstrated that an interaction existed between the participants’ behavior and the examiners’ expectations (Doyen et al., 2011). However, Bargh et al.’s (1996) finding that unconscious behavioral priming employs different cognitive processes than what was previously thought is consistent with Doyen et al.’s (2011) hypothesis; that is, the priming effect may be primarily driven by the experimenter’s expectations rather than just the presence of words. The same was reported in another study, where it was hypothesized that affective priming facilitated general attentional orienting. Rather, the researchers found that the facilitation was more specific to social cues such as eye gaze (Ishikawa et al., 2021). This suggests that the examiners’ expectations, expressed via nonverbal, social cues, confounded the effects of priming in Masling (1965).
A Replication Crisis
Recent research has made great strides in increasing our understanding of priming effects, however, it has revealed an additional surprise for why despite these discoveries, little is still known about the relationship between priming and expectation: Much of that research may contain design flaws (Pashler & Harris, 2012). Therefore, the problem that psychology faces with the issue of solving between priming effects and expectation is a matter of not only is more research needed, but previous research needs to be revisited. Unfortunately, the former is often favored over the latter.
There are a number of reasons why a replication crisis exists in the field of psychology. Occasionally, research may contain falsified results. If a study containing falsified results is replicated, that falsification may then become apparent when the results of the replication do not match the original. Additionally, in studies that have small sample sizes, it may be that a statistically significant result occurred by chance rather than being an accurate reflection of nature. Replicating a study with a small sample size would question the validity of the original if the replication uses a larger sample size than the original did. Lastly, another reason for non-replication may be due to a waning interest in the subject matter itself. The culture of academia often falls victim to fads just as much as any other human industry. The non-replication of a study is often just as likely due to having been forgotten in the midst of a newer, more exciting phenomenon in need of studying as it could be due to being falsified. Research is hard to publish. The newer and shinier the subject matter, the more incentive there is for publication — both for the researchers as well as the publishers.
Some journals have a tendency to favor significant effects, creating a welcome environment for false positive findings. Also known as a publication bias, these journals may encourage researchers to do whatever means necessary for publication, including altering the results of their studies.
Because of an academic culture and publication incentive that rewards conceptual replication rather than actual replication of studies, researchers know that falsification may go unnoticed since conceptual replication is often favored over direct replication. But where the former takes precedence over the latter, scientists can often veer from reality. This worsens both academic trends for shocking information and publication bias by removing the very thing that keeps them in check. Whether falsified or unintentionally flawed, poor research may be prevented by encouraging more direct replication of studies so that it discourages researchers from falsifying information in the first place. On the other hand, however, conceptual replication helps confirm that what one thinks is being tested is really what’s being tested by doing so in a variety of ways. According to a study by Open Science Direction, it was found that correlational tests suggest that replication success was better predicted by the strength of original evidence than by how closely those studies were replicated, (Open Science Collaboration, 2015). So, while direct replication is certainly important to the integrity of the academic community, one runs the risk of continuing to make the same mistakes over and over again if one relies only on reproducing the same results.
This Study Needs To Be Replicated
Masling (1965) is one of many studies that is in dire need of replication to account for experimental design flaws, such as the use of a small sample size of participants, the absence of a control group, the use of a tape recorder rather than a video camera, disregarding responses that may have been outside the definition of a human or animal response (e.g., tree, fear, etc.), and the nonrandom assignment of male examiners to male test-takers and female examiners to female test-takers. Any one these practices may have led to a Type I error (a false positive effect), and could have major implications for how indoctrination is understood.
Overall, implications for this research may not only aid in the understanding of how priming effects interact with examiner expectation for the improvement of test proctoring, but also for ensuring justice practices in human interrogation, combatting misinformation, and promoting critical thinking in the classroom. For instance, if an interrogator is already aware of a particular event or piece of evidence, and he or she is attempting to elicit information from a suspect, the interrogator might lead the suspect to provide that information whether that information is true or not. How the spread of misinformation may be amplified by the medium and/or personality it is presented may also be illuminated. Lastly, priming effects and examiner expectation may also affect critical thinking in the classroom, where an instructor may accidentally rob his or her students from arriving at a particular answer on their own. While these are only a few examples, the applicability of this research will surely have important implications for social cognition, cognitive development, and education research overall.
American Psychological Association. (n.d.). Priming. Retrieved February 05, 2021, from https://dictionary.apa.org/priming
Bargh, J. A., Chen, M., & Burrows, L. (1996). Automaticity of social behavior: Direct effects of trait construct and stereotype activation on action. Journal of Personality and Social Psychology, 2, 230–244. https://doi.org/10.1037/0022–3514.71.2.230
Choca, J. P., & Rossini, E. D. (2018). Assessment Using the Rorschach Inkblot Test. American Psychological Association.
Doyen, S., Klein, O., Pichon, C.-L., & Cleeremans, A. (2011). Behavioral priming: It’s all in the mind, but whose mind? PLoS One, 7(1), e29081. https://doi.org/10.1371/journal.pone.0029081
Graf, P., Shimamura, A. P., & Squire, L. R. (1985). Priming across modalities and priming across category levels: extending the domain of preserved function in amnesia. Journal of Experimental Psychology: Learning, Memory, and Cognition, 11(2), 386–396. https://doi.org/10.1037/0278–7393.11.2.386
Ishikawa, M., Haensel, J. X., Smith, T. J., Senju, A., & Itakura, S. (2021). Affective priming enhances gaze cueing effect. Journal of Experimental Psychology: Human Perception and Performance, 47(2), 189–199. https://doi.org/10.1037/xhp0000880
Masling, J. (1965). Differential indoctrination of examiners and Rorschach responses. Journal of Consulting Psychology, 29(3), 198–201. https://doi.org/10.1037/h0022120
Musen, G., & Squire, L.R. (1992). Nonverbal priming in amnesia. Memory & Cognition, 20(4), 441–448. https://doi.org/10.3758/BF03210928
Open Science Collaboration. (2015). Estimating the reproducibility of psychological science. Science, 349(6251). https://doi.org/10.1126/science.aac4716
Pashler, H., & Harris, C. R. (2012). Is the replicability crisis overblown? Three arguments examined. Perspectives in Psychological Science, 7, 531–536. https://doi.org/10.1177/1745691612463401
Pell, M. D. (2005). Nonverbal emotion priming: evidence from the ‘facial affect decision task’. Journal of Nonverbal Behavior, 29(1), 45–73. https://doi.org/10.1007/s10919–004–0889–8
Schwartz, B. L. (2018). Memory: Foundations and Applications (3rd ed.). SAGE Publications.