Sports Expertise

Expanding the benefits of expertise

Published

October 2018

Team Members

Xiaoping Fang, Deepan Kamaraj, Elizabeth Haley, Marc Coutanche

Context

LeNS Lab, University of Pittsburgh

Role

I served as first-author for this published manuscript. The team developed the research methods and contributed to data analysis together. Another team member and I created our Qualtrics surveys and led online data collection. Finally, I drafted the manuscript.

Learn More

To learn more, you can read the full publication or listen to a podcast episode featuring this project.

Summary

People who develop expertise in a certain domain are known to be able to learn new information in that domain better than those without as much experience. In this experiment, we wanted to know just how powerful this expertise was - could it aid in learning information outside of their domain expertise? We found that expertise might be able to behave as a mnemonic device. Simply link what you are learning to the domain and, tada (!), enhanced memory.

THE PROBLEM

Expertise can give people an advantage when learning in the domain of their expertise. For example, chess players can learn and remember layouts of chess boards even if they are only allowed short glimpses of the board AND even if the pieces are in illegal placements. Not many have looked at another area of memory, though - the area of semantic memory, responsible for maintaining facts. When we created this study, we wanted to know if expertise could support this area of memory as well even when the learners had no idea their memories were about to be tested. In addition, we wanted to know if we could push the limits of this learning advantage. If experts did have an advantage for learning new facts, we wanted to know if they could also have an advantage for learning totally unrelated information, so long as it was associated with related information while they were learning.

THE PROCESS

Participants

We collected data from 130 participants using Amazon Mechanical Turk using a survey created in Qualtrics. Ultimately, we judged that 38 of our participants were sports experts and 34 were non-sports experts.

Expertise Questionnaire

For this study, we focused our attention on sampling sports knowledge experts, as they are relatively common in the population. In order to measure people’s perceptions of their own expertise, we asked participants to complete a self-report questionnaire using a 7-point Likert scale to judge 4 statements like, “I feel competent about my knowledge in sports”; “Compared to others, I know less about sports.”

Learning Task

Because we were interested in how experts learned information unrelated to their expertise, we had participants learn the **faces of unknown people**. Participants viewed the images of 40 men, 20 of whom were implied to be athletes and 20 of whom were implied to be non-athletes. While viewing these men, participants quickly answered **yes/no questions** about the physical features of the men. Importantly, embedded in these questions was the information about the men’s athlete status - we never explicitly asked participants to pay attention to the job application titles of the men they were judging.

Filler Task

To ensure that participants were not rehearsing what they had just learned, we asked for them to complete a 5-minute English vocabulary test. Even though they were never told they would be asked to remember the earlier information, this short distraction gave us the reassurance that no one was actively thinking about the faces or job application titles they had just been exposed to.

Memory Tests

We tested participants’ memories using two main tests.

Old/New Judgment Test - Participants saw 40 images of men, half of which they had seen earlier and half of which were totally new. For images they rated as “old,” participants were asked to describe the job application title of the man in question.
Multiple Choice Test - Over the course of 20 trials, participants viewed 3 images of men at a time along with a job application title. One face matched the description, one did not but was the same condition (i.e., if the match was an athlete, a different athlete), and did not and was from the other condition (i.e., if the match was an athlete, a non-athlete). Participants were tasked with judging which face matched the job application title listed and then rating how confident they were in their judgment on a scale of 1 (not confident) to 4 (very confident).

Sports Knowledge Test

In order to ensure our sports experts truly held extensive knowledge of sports, all participants were allotted 4 minutes to complete a 15-question knowledge test covering football, baseball, and basketball.

THE DATA ANALYSIS

Assessing Expertise

In order to ensure that there were statistically significant differences in sports knowledge between our sports experts and non-sports experts, we examined our participants’ responses to the expertise questionnaire and sports knowledge test.

	Sports-experts	Non-sports-experts
Expertise questionnaire	4.67 (.94)	3.56 (.44)
Sports knowledge test	.80 (.10)	.26 (.10)

Note. Means reported with standard deviations in parentheses.

We tested this difference using an independent t-test (A/B test). Sports-experts performed significantly better in the sport knowledge test than non-sports-experts, t(70) = 23.65, p < .001, d = 5.60. Group differences were also reflected in the self-report expertise questionnaire, t(70) = 6.33, p < .001, d = 1.52.

Old/New Test

Once we had confirmed there were differences in sports expertise between our groups, we began examining memory performance for the faces and the job application titles. We assessed this using 2 (Expertise: sports-expert, non-sports-expert) × 2 (Descriptor: athlete, non-athlete) mixed ANOVAs.

Relative to non-sports-experts, participants with superior sports expertise were able to better recall job application titles associated with each face.

The groups were similarly familiar with the images when the faces were prospective athletes or job candidates.

Multiple-Choice Test

Participants who were sports experts could better select associated faces when presented with two other familiar, but incorrect faces.

they were also more confident when making their judgments.

Not only were experts more accurate at doing this…

THE TAKEAWAYS

Expertise does support memory for information unrelated to the domain of expertise so long as it is associated with the area of expertise when it is learned.
Interestingly, this seems to apply to associations between related words (job titles) and unrelated images (unknown faces) but not to the images alone.