Investor cognitive engagement with earnings information: evidence from pupillary response

Abstract

Using pupillary response data collected in a laboratory setting, this study examines the antecedents and consequences of investors’ cognitive engagement with earnings press releases. We predict that investors’ cognitive engagement is higher when key financial information is included in the headline section of the earnings release, and that their level of cognitive engagement affects their willingness to invest. Our experiment reveals that when reading an earnings release that headlines key financial information, investors exhibit enhanced cognitive engagement as evidenced by increased pupil dilation compared to when reading a release that does not include key financial information in the headline section. These results hold for both good and bad earnings news. Furthermore, OLS regressions show that investors with greater pupil dilation, on average, exhibit increased willingness to invest only when the earnings news is positive. This relation is however nonlinear, with a significant impact on investment decisions observed only when cognitive engagement is relatively high. We also find that investors’ cognitive engagement varies when they read different sections of the earnings release. Taken together, our findings use pupillary responses to provide new insights into investors’ cognitive reactions to earnings news.

Keywords:

• pupillary response
• cognitive engagement
• earnings press release
• headline
• disclosure

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplemental data

Supplemental data for this article can be accessed here http://dx.doi.org/10.1080/00014788.2024.2344155.

Notes

1 The disclosure content in earnings announcements is discretionary, except for the guidance on firms’ disclosure of non-GAAP financial measures under Regulation G. In addition, the way that the earnings releases are written and formatted is discretionary, other than the SEC’s ‘plain English’ requirement.

2 See Medicine LibreTexts: 14.1A: Comparing the Somatic and Autonomic Nervous Systems, https://med.libretexts.org/.

3 We exclude 12 participants from our original sample of 176 participants as their pupillary response data was not captured by the eye tracker in the middle of the session due to technical reasons. We further exclude 11 additional participants whose initial score on their willingness to invest was missing. None of our participants wore glasses during the experiment.

4 Tapestry, Inc. is a modern luxury accessories and lifestyle brand that is the parent company of the Coach, Kate Spade New York, and Stuart Weitzman brands. Tapestry, Inc. is traded on the New York Stock Exchange.

5 The second computer used to track eye movements is not connected to the internet, and therefore cannot be used to collect survey data from Qualtrics.

6 The demographic questions in the experiment survey are multiple choice questions, where the possible answers that participants select from are composed of ranges, except for the question about gender.

7 When aiming at an area-of-interest to extract information, human eyes do not remain completely stationary. Instead, they engage in continuous miniature movements, i.e., ocular drift, to bring the object into the center of visual fixation (Fisher et al. 2011, Krauzlis et al. 2017). To account for such ocular drifts, eye tracking studies often combine consecutive gaze points on the same area-of-interest into one eye fixation (e.g., Myers and Schoelles 2005, Privitera and Stark 2000, Shi et al. 2013). We follow this procedure in processing the raw gaze point output and average pupil sizes.

8 Specifically, the mean responses are 80.388 and 27.738 to question (1), 82.676 and 13.094 to question (2), and 89.603 and 34.082 to question (3) for the good news and bad news conditions, respectively.

9 We also conduct a regression analysis by categorizing pupil size observations into quartiles. The result remains consistent, i.e., only pupil size in the top quartile significantly impacts the percentage change in willingness to invest (β = 3.406, p-value = 0.081, two-tailed) in the good news condition; and none of the quartile indicator variables are significant in the bad news condition.

10 In an untabulated analysis, we examine whether headlining key financial information affects the relation between investors’ cognitive engagement and their investment decisions. Specifically, we regress %Change in Willingness to Invest on Pupil size, HeadliningKeyFinInfo, as well as the interaction between the two. The results reveal that in the good news condition, Pupil size significantly increases the willingness to invest (two-tailed p-value = 0.021) while the coefficients on HeadliningKeyFinInfo (two-tailed p-value = 0.939) and Pupil size × HeadliningKeyFinInfo (two-tailed p-value = 0.130) are both insignificant. In the bad news condition, on the other hand, none of the independent variables are significant.

Additional information

Funding

This work was supported by Santa Clara University Research Grant: [Grant Number GR101142].