Relational Responding Task (RRT)

Background

The Relational Responding Task (RRT) is a cognitive-behavioral test designed by Jan De Houwer and colleagues in 2015 to measure implicit beliefs with the potential to capture differences in how cognitively paired categories are related to each other. For example, the categories "I" and "physical exercise" might be closely related concepts in my mind, but the underlying relationship might be 'need to' vs. 'want to' which can have different consequences for my actual workout behavior. The general procedure of the RRT as well as its score calculation algorithms borrow heavily from the popular Implicit Association Test (IAT). Furthemore, like the IAT, the RRT is simple to set up and run and tends to have a low drop-out rate (which can be a problem for the more complex and longer Implicit Relational Assessment Procedure (IRAP)). Unlike these more traditional implicit belief tests, however, the RRT presents concepts as complete statements (e.g., "I am a fun person to hang out with") to capture the underlying relationships between concepts.

The RRT uses latency measures and forces people to categorize actual statements according to pre-determined rules that they may or may not agree with. Between blocks, the rule changes to the opposite one (e.g. from 'act as if boys are smarter than girls' to 'act as if girls are smarter than boys') and the difference in response behavior under these two opposing rules is used as a measure of a person's implicit biases and beliefs.

The Inquisit RRT template script uses the classic flower-insect categories with statements such as "flowers are prettier than insects" but the categories and statements can easily be customized.

Task Procedure

The procedure of the RRT looks similar to that of the IAT and is divided into 7 blocks. Participants first practice the 'true' and 'false' response keys by categorizing alternative 'inducer' words for 'true' (e.g. 'accurate') and 'false' (e.g. 'faulty') using the Q-key for 'true' and P-key for 'false'. This practice block (Block1) runs 20 trials total (10 inducer stims for 'true' and 10 inducer stims for 'false', in random order). The inducer stimuli are presented in orange on a black background throughout the RRT procedure.

The following 6 blocks are divided into a 'Rule1' testing round (one 'Rule1' practice blocks, followed by two test blocks) and a 'Rule2' testing round (one 'Rule2' practice block, followed by two test blocks). Rule1 and Rule2 are opposing rules with Rule1 asking participants to respond as if 'flowers are more beautiful than insects' and Rule2 asking them to to respond the opposite way as if 'insects are more beautiful than flowers'. Each practice block runs 10 rule-consistent (should result into a 'true' categorization) and 10 rule-inconsistent (should result into a 'false' categorization) statements in random order. The test blocks run the 10 rule-consistent and 10 rule-inconsistent statements in a random order, interspersed with the 20 inducer stimuli from the very first practice block. The statements are presented one-by-one (with inducer stimuli in orange and target statements in blue), and participants are asked to respond as quickly as possible without making too many mistakes. If a mistake is made, error feedback is provided in the form of a red 'X' and the response has to be corrected.

Standardized mean difference scores (d-scores) are calculated for the two testblocks of the two rule testing rounds (the first test block of the Rule1 testing round is compared to the first test block of the Rule2 testing round; ditto for the second test blocks), and their unweighted mean is taken as the overall d-score. Positive d-scores suggest a bias to finding flowers more beautiful than insects; negative d-scores suggest the opposite.

What it Measures

The Relational Responding Task (RRT) measures implicit biases

Psychological domains

• Implicit Cognition: Cognitive Processes not under conscious control
• Social Cognition: People's thoughts and perceptions of themselves and others in the social world

Main Performance Metrics

• d-score: Measure of strength between associated concepts

Psychiatric Conditions

The RRT has been used with the following patient groups

• Depression
• Anxiety
• Substance Abuse
• Eating Disorders

Test Variations

References

Search Google Scholar for peer-reviewed, published research using the Inquisit Relational Responding Task (RRT).

---

De Houwer, J., Heider, N., Spruyt, A., Roets, A. and Hughes, S. (2015). The relational responding task: toward a new implicit measure of beliefs. Frontiers in Psychology. 6: 319.https://doi.org/10.3389/fpsyg.2015.00319.

Rosseel, L., Speelman, D., & Geeraerts, D. (2019). The relational responding task (RRT): a novel approach to measuring social meaning of language variation. Linguistics Vanguard : Multimodal Online Journal, 5(1). https://doi.org/10.1515/lingvan-2018-0012

Fida, R., Ghezzi, V., Paciello, M., Tramontano, C., Dentale, F., & Barbaranelli, C. (2022). The Implicit Component of Moral Disengagement: Applying the Relational Responding Task to Investigate Its Relationship With Cheating Behavior. Personality & Social Psychology Bulletin, 48(1), 78–94. https://doi.org/10.1177/0146167220984293

Dentale, F., & Vecchione, M. (2025). Implicit Measures of Agentic Narcissism and Their Relationships with Self-Enhancement. Journal of Personality Assessment, 1–13. https://doi.org/10.1080/00223891.2025.2509495