Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Dec 1.
Published in final edited form as: J Clin Exp Neuropsychol. 2024 Nov 23;46(10):913–922. doi: 10.1080/13803395.2024.2432655

Initial Expressed Emotion During Neuropsychological Assessment: Investigating Motivational Dimensions of Approach and Avoidance

Karlee Patrick a, Erin Burke a, John Gunstad a, Mary Beth Spitznagel a
PMCID: PMC11802313  NIHMSID: NIHMS2037509  PMID: 39579339

Abstract

Objective.

Prior work indicates that discrete emotions are linked to performance across multiple domains of cognitive function, and thus have the potential to impact cognitive profiles in neuropsychological assessment. However, reported presence and magnitude of the relationships between emotion and cognitive test performance are inconsistent. Variable findings in this regard could be due to failure to consider motivations associated with expressed emotion. To better understand the potential impact of expressed emotion on neuropsychological test performance, it may be beneficial to consider approach and avoidance motivation during assessment.

Method.

The current cross-sectional study examined associations between cognitive performance and digitally phenotyped facial expressions of discrete emotions on dimensions of approach (i.e., joy, sadness, anger) and avoidance (i.e., fear, disgust) in the context of virtual neuropsychological assessment in 104 adults (ages 55–90).

Results.

Initial facial expressions categorized as anger and joy predicted later reduced cognitive performance in aspects of memory and executive function within the virtual session, respectively. Test performance was associated neither with sadness nor with avoidance emotions (i.e., disgust or fear).

Conclusions.

Results of the current study did not strongly align with approach/avoidance explanations for links between emotion and cognitive performance; however, results might support an arousal-based explanation, as joy and anger are both high arousal emotions. Additional investigation is needed to understand the intersection of emotion motivation and physiological arousal in the context of neuropsychological assessment.

Keywords: neuropsychological assessment, emotion, facial expressions

INTRODUCTION

The expression of emotion consists of complex biological, learned, and evolved processes that influence all parts of life, including cognitive functioning (Tyng et al., 2017). Emotion has been linked to multiple domains of cognitive function, including executive functions (e.g., problem solving, abstract reasoning), attention, learning, and memory, particularly encoding and retrieval of information (Tyng et al., 2017; Jung et al., 2014). However, prior investigations of relationships among cognitive function and discrete emotions (i.e., joy, anger, sadness, fear, disgust) produced divergent results suggesting that emotions may impair, enhance, or exert no influence on cognitive function at all (Allen et al., 2014; Kalanthroff et al., 2013; Padmala et al., 2011; Driesbach, 2006; Shields et al., 2016). The potential impact of emotion on cognitive test performance is important for the neuropsychologist to understand, as it may inform interpretation of a cognitive profile.

As noted, past research has yielded mixed findings. Some work suggests that anger may impair memory and executive functions, particularly decision making (Meissner et al., 2021; Garfinkel et al., 2016; Threadgill & Gable, 2019), while others find no impact (Shields et al., 2016). Sadness has been associated with improved memory performance and problem solving (Threadgill & Gable, 2019; Bonanno et al., 2008), impaired executive function (Beaulieu-Pelletier et al., 2021), but was unrelated to working memory performance (Storbeck & Maswood, 2016). Fear has been associated with impairments in attention, executive functions (particularly cognitive inhibition), processing speed, and reaction time (Garfinkel et al., 2016; Lindström & Bohlin, 2012). Disgust may be associated with impairments in processing, memory, and attention in disgust-relevant situations (Knowles et al., 2019). Regarding positive emotions, joy may enhance working memory performance and executive function, including shifting between demands of a task (Storbeck & Maswood, 2016; Yang et al., 2013; Wang et al., 2017). The degree to which prior research findings would generalize to neuropsychological assessment is not certain. Previous studies have relied heavily on undergraduate samples and induced emotion manipulations (e.g., using video clips, emotional images, or autobiographical emotion memory tasks; Storbeck & Maswood, 2016; Yang et al., 2013; Wang et al., 2017). Whether similar associations would be found in naturally occurring (i.e., non-induced) emotion states or in other age groups remains unclear.

The lack of a clear and robust pattern of relationships between emotion and cognitive test performance in prior research may also stem from failure to consider emotion motivation (Roseman, 2013; Shields et al., 2016). One common conceptualization for studying emotion motivation is the approach versus avoidance model, in which the individual must determine how to respond to new situations based on their appraisal of that event (Roseman, 2013). A situation can be appraised as involving appetitive (approach) or aversive (avoidance) motivations; these motivations, in turn, drive emotion responses (Roseman, 2013; Coifman, forthcoming). Emotions on the approach dimension include anger (motivation to approach a challenge after a personal goal is blocked; Carver & Harmon-Jones, 2009), sadness (motivation to seek social support after a loss; Bonanno et al., 2008), and joy (motivation to seek social rewards; Roseman, 2013). In contrast, emotions on the avoidance dimension include fear and disgust, which motivate a person to avoid situations that could cause physical and/or psychological harm (Roseman, 2013; Coifman, forthcoming). Approach and avoidance motivations may be associated with unique autonomic nervous system responses, patterns of brain activity, and neurotransmitter activity that may be associated with different impact on cognitive function (Shields et al., 2016; Moons & Shields, 2015; Kreibig, 2010). The motivational orientation of expressed emotion may thus be important to consider in the context of neuropsychological assessment.

To expand upon prior research, the current study explored associations among cognitive test performance and facial expressions of discrete emotions in the context of virtual neuropsychological assessment in a sample of adults aged 55–90. Expressed emotions on motivational dimensions of approach and avoidance were examined using digital phenotyping of facial expressions at session onset to capture approach/avoidance orientation to neuropsychological assessment and investigate whether participants’ initial orientation to the assessment predicted cognitive performance throughout testing. Given the lack of robust patterns in links between emotion and cognitive performance in prior work, and that previous research did not examine these associations in the context of neuropsychological assessment, relationships were explored without formal directional hypotheses.

METHOD

Participants

Community-dwelling adults between the ages of 55 and 90 were recruited. Participant recruitment materials stated that the purpose of the current study was to investigate cognitive testing in a virtual, computer-based format. A total of 116 participants took part in the virtual study. Inclusion criteria required that participants were able to speak/read English, had the ability to use a computer and teleconferencing, had sufficient internet access to complete testing, and had sufficient visual and auditory abilities to interact with the examiner. Exclusion criteria included insufficient video quality for analysis, invalid performance as detected by the Reliable Digit Span (see below for more detail), or incomplete data on the below variables.

Measures

Three primary sources of data were used in the current study: demographic information, facial expressions of emotion, and measures of cognitive functioning.

Demographic Information

Participants were asked to self-report their age, gender, education, and race/ethnicity.

Measurement of Emotion Expression

Participants’ facial expressions of emotion were measured using a digital phenotyping software package, OpenDBMv2.0 (AiCure, 2021). Automatic coding of facial expressions of emotion presents a well-validated alternative to self-report of emotion that improves objectivity of measurement, addresses self-report bias, and is less disruptive for the participant than other methods (Shen et al., 2022). The software included a measurement of emotion intensity for each discrete emotion based on a combination of Facial Action Coding System (FACS; Ekman & Frieson, 1978; Ekman, 1992) action units associated with each emotion. FACS has been extensively validated in prior literature (AiCure, 2021; Sayette et al., 2001; Clark et al., 2020; Ekman & Frieson, 1978; Ekman, 1992). OpenDBMv2.0 software has been validated as a digital, automatic measurement of daily, naturalistic facial expressions of emotion in the context of major depressive disorder (Abbas et al., 2021; Schultebraucks et al., 2022) and schizophrenia (Abbas et al., 2022; Galatzer-Levy et al., 2020) but has not yet been employed in the context of neuropsychological assessment. This software was designed to be robust to individual differences in appearance (e.g., glasses, wearing a hat) and included a large, diverse sample when developing an algorithm for automatic coding of facial expressions of emotion (AiCure, 2021) For each separate discrete emotion, a continuous vector (ranging from 0–1) that measured intensity of facial expression of emotion was generated and used in analyses. An intensity value above zero was only identified by the software when all facial action units associated with that discrete emotion were present (AiCure, 2021).

Assessment of Cognitive Function

Multiple domains of cognitive function were assessed. Assessment focused on executive function, attention, learning, and memory, as these domains have been most consistently associated with emotion expression in prior work.

Verbal Fluency.

Participants were asked to verbally generate words that begin with the letters C, F, and L in three separate, one-minute trials for each as a test of phonemic fluency. The total number of C, F, and L words generated were summed and normed into a T-score. This overall T-score for CFL fluency was entered in analyses. Verbal fluency tasks show high sensitivity and specificity in identifying cognitive impairments in adults (Lezak et al., 2004).

Cognitive Set Shifting.

Participants’ cognitive set shifting ability was assessed using the Oral Trail Making Test B (OTMT-B; Ricker & Axelrod, 1994). Participants were asked to count aloud, switching between number and letter (i.e., 1-A-2-B etc.), until they reached number 13. The test was discontinued if the participant had not completed the task in 5 minutes or made 5 errors. This measure removes motor and visual demands of the original written version and is moderately correlated (r-.62; Mrazik et al., 2010) with the written version of the Trail Making Test developed by Reitan (1955). Participant times on Oral Trail Making Test B were normed into T-scores based on clinical norms, and scores were coded such that lower T-scores (slower completion) reflected worse performance.

Immediate and Delayed Recall.

Episodic memory (immediate and delayed) was assessed using the Craft Story Immediate Recall and Craft Story Delayed Recall (Craft Story 21; Craft et al., 1996). Participants received points for both verbatim and paraphrased recall which were summed and entered separately in analyses as normed T-scores. Craft Story 21 demonstrates sound psychometric properties (Weintraub et al., 2018), including reliability (intraclass correlation coefficient = 0.77 [verbatim] and 0.81 [paraphrase]; Howard et al., 2023) and convergent validity with other measures of episodic memory (Monsell et al., 2016).

Auditory Attention and Working Memory.

Participants’ auditory attention was assessed using the Digit Span Forward test. Digit Span Backward and Digit Span Sequencing assessed participants’ auditory attention, working memory, as well as executive function skills. Scores were summed, normed, and entered separately in analyses. Digit Span is well-validated and shows excellent reliability (Lichtenberger & Kaufman, 2013). Reliable Digit Span (RDS; Greiffensten, Baker, & Gola, 1994), a well-established measure that is embedded in the Digit Span test, was also calculated to assess performance validity.

Procedure

Study procedures followed Strengthening and Reporting of Observational Studies in Epidemiology (STROBE) guidelines for cross-sectional studies (von Elm et al, 2014) and were approved by the Kent State University Institutional Review Board. Data collection occurred from 07/14/2021 to 12/15/2022. Participants provided informed consent for research including video recording of the session. Participants were told during the consent process before beginning the study that they would not receive individual feedback on their cognitive functioning based on results of the study. After providing consent, participants completed a 60-minute neuropsychological testing session through Microsoft Teams. This session included the above-described cognitive tests as well as self-report questionnaires (i.e., demographic information). Participants were offered a $25 gift card for taking part in the study. The study session was recorded for each participant, and videos were split into clips for subsequent analysis. One-minute video clips of participants completing a semantic fluency test at the beginning of the session were entered into OpenDBM 2.0 software for automatic coding of facial expressions of emotion.

Statistical Power

Prior research exploring the influence of induced emotions on cognitive function indicated small effects (partial eta square = .10; Padmala et al., 2011). A priori power analyses conducted using G*Power (Faul et al., 2009) with an alpha of .05 and a power of .80 suggested a minimum sample size of 96 is needed for correlation and regression-based analyses to detect effects at the low end of effect size ranges in prior research. Intentional over-recruiting was conducted to ensure an adequate sample size in the event of incomplete data or insufficient video quality.

Analytic Strategy

All analyses were conducted with SPSS version 29. Descriptive statistics were conducted for primary variables to determine acceptability for parametric analyses. Pearson correlations bootstrapped with 5,000 samples were used to examine associations among facial expressions of emotion and neuropsychological test results. Neuropsychological tests that correlated significantly with facial expressions of emotion based on these Pearson correlations were then entered as dependent variables in separate linear regressions. Demographic variables that correlated significantly with test performance or facial expressions of emotion were entered in the first step, and facial expressions of emotion were entered in the second step in separate linear regressions. Familywise α was set at 0.05 with Holm–Bonferroni adjustment applied to account for multiple analyses and minimize type I error.

RESULTS

Preliminary Analyses

Measures of all facial expressions of emotion were non-normally distributed and positively skewed (skewness >2.0 and kurtosis >6.0). Due to non-normality, analyses using these variables were confirmed using bootstrapping procedures with 5,000 bootstrapping samples (Field, 2020). All other variables met normality assumptions. Participant education and race/ethnicity correlated significantly with neuropsychological test performance and were included as control variables in linear regression analyses.

Participants

A total of 8 participants were excluded for poor video quality or insufficient data due to experimenter error. A further 4 participants were excluded for having an RDS of 7 or lower (Schroeder et al., 2012), indicating poor effort on neuropsychological testing. The final sample included 104 participants. Participants were 69.5 years of age (SD = 7.4) on average, and well-educated, with an average of 15.7 years of education (SD = 2.4). Participants were largely white (94.2%) and female (67.6%). Performance on neuropsychological tests was average, with T-scores ranging from 46 to 55. Please see Table 1 for descriptive statistics of sample demographics and Table 2 for descriptive statistics of primary variables.

Table 1.

Descriptive statistics for primary variables (M/SD,range)

Participant Demographics n = 104
Age (M/SD/range) 69.5/7.4/55–90
Gender (%)
Male 32.4%
Female 67.6%
Education (M/SD/range) 15.7/2.4/12–21
Race/Ethnicity (%)
White/Caucasian 94.2%
Black/African American 4.7%
Asian 0%
Hispanic 1.0%
Other 1.1%
Open in a new tab

Note. M=mean; SD=standard deviation

Table 2.

Descriptive statistics for primary variables

Measurement of Emotion Expression
(Vector Scale Ranging 0–1)		 M/SD/range Number of Participants Demonstrating Each Emotion
Facial Expressions of Joy Intensity 0.04/0.07/0–0.43 n = 98
Facial Expressions of Anger Intensity 0.01/0.02/0–0.09 n = 73
Facial Expressions of Sadness Intensity 0.01/0.01/0–0.08 n = 57
Facial Expressions of Fear Intensity 0.001/0.002/0–0.02 n = 25
Facial Expressions of Disgust Intensity 0.01/0.01/0–0.04 n = 81
Neuropsychological Testing (T-Scores)
Craft Story Immediate Verbatim Recall 46.6/9.7/25–70
Craft Story Immediate Paraphrase Recall 46.9/10.0/19–65
Craft Story Delayed Verbatim Recall 46.1/8.9/26–66
Craft Story Delayed Paraphrase Recall 47.1/9.8/22–74
Phonemic Fluency (CFL) 51.5/9.9/24–71
Digit Span Forward 49.9/7.8/31–66
Digit Span Backward 53.5/8.6/37–82
Digit Span Sequencing 53.7/8.5/26–74
Oral Trail Making Test B 55.6/7.9/20–66
Open in a new tab

Note. M=mean; SD=standard deviation

Facial Expressions of Emotion and Neuropsychological Test Performance

Approach-related Expressions of Emotion

Joy.

Higher ratings of expressed joy were negatively correlated with performance on Digit Span Sequencing (r = −.19, p < .01 with Holm-Bonferroni adjustment). All other associations were non-significant. See Table 3. Higher ratings of expressed joy significantly predicted lower Digit Span Sequencing performance in linear regressions controlling for participant education and race/ethnicity in the first block (ΔR2 = .03, ΔF(3, 89) = 7.18, p < .01 with Holm-Bonferroni adjustment).

Table 3.

Bootstrapped Pearson correlations between measures of facial expressions of emotion and neuropsychological test performance controlling for participant education and race/ethnicity with 95% confidence intervals.

Neuropsychological Test Facial Expression of Joy Facial Expression of Anger Facial Expression of Sadness Facial Expression of Fear Facial Expression of Disgust
Craft Story Immediate Verbatim Recall 0.08
(−.12, .28)		 −0.15
(−.37, .08)		 0.11
(−.12, .33)		 −0.12
(−.47, .27)		 −0.11
(−.30, .10)
Craft Story Immediate Paraphrase Recall 0.04
(−.17, .24)		 −0.09
(−.31, .15)		 0.16
(−.07, .37)		 −0.01
(−.38, .36)		 −0.04
(−.24, .16)
Craft Story Delayed Verbatim Recall 0.07
(−.14, .27)		 −0.22*
(−.38, −.11)		 0.01
(−.22, .23)		 −0.24
(−.53, .09)		 −0.09
(−.28, .11)
Craft Story Delayed Paraphrase Recall 0.06
(−.15, .26)		 −0.20*
(−.37, −.01)		 0.03
(−.20, .26)		 −0.14
(−.45, .20)		 −0.05
(−.25, .15)
Phonemic Fluency (CFL) −0.09
(−.29, .12)		 0.12
(−.12, .34)		 −0.09
(−.31, .14)		 −0.05
(−.38, .29)		 −0.03
(−.23, .17)
Digit Span Forward −0.10
(−.15, .22)		 −0.03
(−.21, .19)		 0.10
(−.07, .26)		 −0.02
(−.39, .33)		 0.15
(−.05, .34)
Digit Span Backward −0.11
(−.24, .18)		 −0.01
(−.23, .21)		 −0.12
(−.34, .12)		 −0.01
(−.33, .32)		 −0.04
(−.23, .16)
Digit Span Sequencing −0.19**
(−.31, −.10)		 −0.01
(−.24, .23)		 −0.04
(−.26, .19)		 −0.09
(−.41, .24)		 0.01
(−.19, .21)
Oral Trail Making Test B 0.08
(−.14, .33)		 0.05
(−.18, .28)		 0.13
(−.04, .28)		 −0.24
(−.53, .09)		 −0.06
(−.26, .14)
Open in a new tab

Note.

*

indicates significance at p<0.05

**

indicates significance at p<0.01

Anger.

Higher ratings of expressed anger were negatively correlated with performance on Craft Story Delayed Verbatim Recall (r = −.22, p < .05 with Holm-Bonferroni adjustment) and Craft Story Delayed Paraphrase Recall (r = −.20, p < .05). The latter was not robust to Holm-Bonferroni adjustment. Other correlations were non-significant. See Table 3. Anger expressions significantly predicted lower Craft Story Delayed Verbatim Recall performance in linear regressions controlling for participant education and race/ethnicity in the first block (ΔR2 = .08, ΔF(3, 89) = 4.01, p < .01 with Holm-Bonferroni adjustment).

Sadness.

Facial expressions categorized as sadness and test performance were not significantly correlated. See Table 3.

Avoidance-related Expressions of Emotion

Fear and Disgust.

Neither expressions categorized as fear nor those categorized as disgust were correlated with test performance; see Table 3.

DISCUSSION

The current study examined associations among facial expressions of discrete emotions and cognitive test performance in the context of virtual neuropsychological assessment in a sample of individuals 55–90 years of age. To examine how approach/avoidance orientation to neuropsychological assessment relates to test performance, emotions on motivational dimensions of approach (i.e., joy, anger, sadness) and avoidance (i.e., fear, disgust) at session onset were explored.

No strong relationships were detected among avoidance-related emotions (i.e., fear, disgust) and neuropsychological test performance. This lack of association runs counter to a recent review showing significant correlations between higher levels of disgust and lower performance on tests of processing speed, memory, and attention (Knowles et al., 2019). However, all works described in that review induced the emotion by showing participants disgusting images. In contrast, the present study examined natural responses to the neuropsychological assessment; it is likely that neuropsychological testing is not a strongly “disgust-relevant” situation. Similarly, the assessments conducted in the present work may not have been sufficiently “fear relevant” to detect significant relationships. It is possible that greater fear would be present during neuropsychological testing in other contexts, particularly if there is greater threat to the self (e.g., testing in the context of neurologic conditions after which an individual may learn they have cognitive impairment). While cognitive testing in any context could induce performance-based negative emotions, participants in the current study were compensated for their time regardless of performance and were not provided cognitive assessment results; thus, stakes of test performance may not have been high enough to naturally induce fear.

Exploration of associations among facial expressions of approach emotions (i.e., sadness, joy, anger) and cognitive test performance also yielded results that did not align with existing approach/avoidance theories of emotion. Although previous findings have been inconsistent, some work did suggest that facial expressions categorized as joy (i.e., approach emotion) could be associated with better cognitive performance and prior work showed positive associations among performance on memory and executive function tasks and self-reported joy (Storbeck and Maswood, 2016; Wang et al., 2017; Yang et al., 2013). Yet, in the present study, negative associations between facial expressions categorized as joy and performance on a complex attention/executive function task (i.e., Digit Span Sequencing) were observed. Relatedly, because anger ostensibly readies one for a challenge (Carver & Harmon-Jones, 2009), it is conceptualized as an approach emotion and might be associated with improved cognitive performance; however, prior findings vary (Shield et al., 2016; Moons & Shields, 2015; Kriebig, 2010). The present results demonstrated significant negative associations between facial expressions categorized as anger and memory performance. Additionally, based on the conceptualization of sadness as an approach emotion in prior work, a positive relationship between facial expressions of sadness and cognitive performance could be expected, though past findings have been inconsistent as well (Bonanno et al., 2008; Threadgill & Gable, 2019). No significant associations among facial expression categorized as sadness and test performance were observed in the present study.

Though the current results generally did not align with approach/avoidance conceptualizations of emotion, they might be explained, particularly in the case of approach emotions, by the impact of arousal on cognitive control. Expressed joy and anger (higher arousal emotions; Harmon-Jones et al., 2016) were associated with reduced cognitive performance, while sadness (lower arousal emotion; Harmon-Jones et al., 2016) was not associated with cognitive performance. High levels of physiological arousal in the context of approach motivations (e.g., joy, anger) have been associated with lower performance on tasks of cognitive control (i.e., attention, executive function) during cognitively demanding tasks (Cudo et al., 2018; Miller & Cohen, 2001). Additionally, the mechanism of action in the relationship between sadness (low arousal emotion) and improved cognitive performance has been hypothesized as decreased arousal leading to improved information processing, attention to detail, and memory performance (Bonanno et al., 2008; Threadgill & Gable, 2019). Associations between arousal and cognitive performance likely operate on an inverted U-shaped curve, with optimal cognitive performance observed at moderate arousal levels, and reduced cognitive performance at high and low arousal levels (Yerkes & Dodson, 1908; Ciria et al., 2021). It is plausible that facial expressions categorized as joy and anger in the current study indicated arousal was elevated enough to interfere with cognitive performance on tasks of executive function and working memory, while facial expressions of sadness were not associated with cognitive performance due to corresponding lower levels of arousal. To further investigate arousal-based explanations, links among fear (high arousal emotion; Harmon-Jones et al., 2016) and cognitive performance will need to be explored in neuropsychological assessments with higher stakes for the individual as described above. Additionally, fully understanding links among approach/avoidance motivation, arousal, and cognitive performance will require capturing of facial expressions and measurement of physiological arousal throughout neuropsychological testing in future work, as expressed emotions and their underlying motivations could shift during the course of the assessment process.

An alternative explanation for some of the present results is that facial expressions categorized as joy by the digital phenotyping software detected “nervous smiles” in the context of neuropsychological assessment. Of note, prior research suggests that individuals experiencing anxiety in social situations may employ expressive dissonance (i.e., smiling when anxious to express the opposite of what one feels) to facilitate social bonds and likeability (Bahl & Ouimet, 2022). Nervous smiling could be an indication of anxiety in the context of neuropsychological assessment, and this anxiety could, in turn, be related to reduced performance. Relatedly, it is possible that facial expressions categorized as anger by the digital phenotyping software instead detected facial expressions of concentration in the context of neuropsychological assessment. Indeed, lowered eyebrows and tightening of eyelids (seen in facial expressions of anger) have been categorized as concentration states in prior work (Rozin & Cohen, 2003). Particularly given that participants with low effort as detected by the RDS were removed from the sample, it is possible that individuals in the current study concentrated harder on tasks they found challenging. Poorer performance may have been yielded for those tasks that participants found most challenging. Whether current results can be explained by nuanced expression of emotion (i.e., nervous smiling vs. joy; concentration vs. anger), the impact of approach/avoidance motivations, and/or levels of physiological arousal will need to be teased apart in future work.

The current study offers several important contributions. Prior research exploring links among emotion and cognitive performance relied almost exclusively upon emotion induced in a laboratory setting. Investigation of these associations had not been previously conducted in the more applied context of a battery of neuropsychological tests, without explicit effort to artificially induce a specific emotion. The current study sought to explore naturally occurring facial expressions of emotion in a previously unstudied context that may more closely mimic actual clinical assessment, with the goal of producing findings with greater ecological validity for neuropsychological evaluation as it is conducted in the field. Further, prior work has explored only limited cognitive domains, and this literature gap was addressed in the current study through inclusion of neuropsychological testing across multiple cognitive domains.

Despite these contributions, several limitations of the current work must also be noted. A primary methodological concern is that facial expressions of emotion observed in the current study may not fully capture the range of emotions expressed during a typical clinical assessment. Results of neuropsychological evaluation often connote significant impact for the lives of individuals undergoing testing and may elicit a great range of emotions (e.g., worry, frustration, fear, relief, sadness, anger; Shaefer et al., 2023; Casaletto & Heaton, 2017; Lai et al., 2009). Although the structure of methods used in the current study was established to closely mimic actual neuropsychological assessment, the stakes of test performance were likely lower for most participants in the current study, as there were no real-life consequences related to the outcome of testing. In addition to this potential limitation, participant recruitment materials stated that the current study was investigating cognitive testing in a virtual, computer-based format. It is possible that individuals who were experiencing subjective cognitive difficulties and were concerned about their memory signed up for the study for that reason, and this may have influenced their emotional expression at the beginning of the session. However, to address this issue, participants were told during the consent process before beginning the study that they would not receive individual feedback on their cognitive functioning based on results of the study. Additionally, many individuals presenting for neuropsychological testing report subjective cognitive concerns that led them to seek cognitive testing, and participants’ reactions to being told they would be doing cognitive testing might mimic reactions expressed by an individual undergoing neuropsychological testing in a clinic context. Factors limiting generalizability of results to broader populations are acknowledged. Participants in the current study were largely white residents of a midwestern community; known racial and ethnic differences in facial expressions of emotion (Fan et al., 2021) could influence generalizability of results across race and culture. Additionally, adults typically presenting for neuropsychological assessment often exhibit some degree of functional and cognitive decline (Casaletto & Heaton, 2017); however, the current sample was highly educated and showed average performance on testing (with restricted range of test scores), perhaps due to cognitive reserve (Wilson et al., 2019; Stern, 2012). They were also able to access virtual, computer-based assessment from their home, suggesting they were not experiencing, on average, functional or cognitive decline to the degree that might typically be observed in an individual seeking neuropsychological assessment.

Findings and limitations of the current study highlight several considerations for further investigation. Future work should explore the intersection of approach/avoidance motivations, high/low levels of arousal, and positive/negative valence (i.e., pleasantness/unpleasantness) of emotions throughout the full course of neuropsychological assessment to tease apart possible explanations for associations among expressed emotions and cognitive performance. Positive/negative valence of emotion expression may be important to consider, as anger, sadness, and fear (emotion expressions with negative valence) correlated with one another in the current study, and expression of joy (emotion with positive valence) was not significantly associated with other emotion expressions (See Table 4 for correlations among facial expressions of emotion). Emotion expressions did not correlate with one another within categories of approach/avoidance emotions, providing further support of the need to consider relationships among approach/avoidance motivations, high/low levels of arousal, and positive/negative valence in future work. Another important consideration is the possibility that emotion expressed in the context of neuropsychological assessment could reflect several underlying factors, including reactions to neuropsychological assessment, longstanding traits or dispositions (e.g., neuroticism, conscientiousness), emotion regulation abilities, and/or recent stressors and experiences. Future work will need to tease apart relationships among these factors that may impact expressed emotion and investigate their differential impact on cognitive performance. Within-subjects emotion induction methods could be used to rigorously investigate associations among cognitive performance, trait-level factors, and state-level emotion expression in the context of neuropsychological assessment as well. Additionally, further examination of the OpenDBM software in the context of naturalistic neuropsychological assessment is needed to ensure that nuanced expressions of emotion such as nervous smiling and confusion are captured. Exploration of results from the current study in the context of clinical neuropsychological assessment or in an assessment with experimental consequences of test performance might better capture emotion expressed during a typical neuropsychological evaluation. Future work should also include individuals with cognitive impairment, those with a more representative range of educational attainment, and greater racial and ethnic diversity.

Table 4.

Pearson correlations between measures of facial expressions of emotion

Facial Expression of Emotion Facial Expression of Joy Facial Expression of Anger Facial Expression of Sadness Facial Expression of Fear
Facial Expression of Anger -0.05 -- -- --
Facial Expression of Sadness -0.09 0.28* -- --
Facial Expression of Fear -0.11 0.17* 0.53** --
Facial Expression of Disgust -0.02 -0.05 0.08 0.10
Open in a new tab

Note.

*

indicates significance at p<0.05

**

indicates significance at p<0.01

Overall, limited support for the influence of approach/avoidance motivations of emotion on neuropsychological test performance was observed. As facial expressions of emotion significantly predicted some aspects of cognitive performance, consideration of emotion expressed at the time of neuropsychological assessment might yet be an avenue for improving validity and accuracy of assessment interpretation. However, expansion of methods used in the present study will be needed to fully understand the potential influence of emotions on the clinical neuropsychological assessment.

Footnotes

Disclosure Statement: This work was supported in part by a National Institute on Aging grant awarded to Dr. John Gunstad (R01AG065432) and by an internal university award granted to Karlee Patrick. There are no conflicts of interest to report.

REFERENCES

  1. Abbas A, Hansen BJ, Koesmahargyo V, Yadav V, Rosenfield PJ, Patil O, Dockendorf MF, Moyer M, Shipley LA, Perez-Rodriguez MM, & Galatzer-Levy IR (2022). Facial and Vocal Markers of Schizophrenia Measured Using Remote Smartphone Assessments: Observational Study. JMIR formative research, 6(1), e26276. 10.2196/26276 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Abbas A, Sauder C, Yadav V, Koesmahargyo V, Aghjayan A, Marecki S, Evans M, & Galatzer-Levy IR (2021). Remote Digital Measurement of Facial and Vocal Markers of Major Depressive Disorder Severity and Treatment Response: A Pilot Study. Frontiers in digital health, 3, 610006. 10.3389/fdgth.2021.610006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. AiCure. (2021). AiCure Makes Code Open Source to Advance Digital Biomarker Development. AiCure. https://aicure.com/ [Google Scholar]
  4. Allen RJ, Schaefer A, & Falcon T (2014). Recollecting positive and negative autobiographical memories disrupt working memory. Acta psychologica, 151, 237–243. 10.1016/j.actpsy.2014.07.003 [DOI] [PubMed] [Google Scholar]
  5. Bahl N, & Ouimet AJ (2022). Smiling won’t make you feel better, but it might make people like you more: Interpersonal and intrapersonal consequences of response-focused emotion regulation strategies. Journal of social and personal relationships, 39(7), 2262–2284. 10.1177/02654075221077233 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Beaulieu-Pelletier G, Bouchard MA, & Philippe FL (2021). Loss-related mental states impair executive functions in a context of sadness. Heliyon, 7(3), e06599. 10.1016/j.heliyon.2021.e06599 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Bonanno GA, Goorin L, & Coifman KG (2008). Sadness and grief. In Lewis M, Haviland-Jones JM, & Barrett LF (Eds.), Handbook of emotions. The Guilford Press. [Google Scholar]
  8. Campbell NM, Dawel A, Edwards M, & Goodhew SC (2021). Does motivational intensity exist distinct from valence and arousal. Emotion, 21(5), 1013–1028. 10.1037/emo0000883 [DOI] [PubMed] [Google Scholar]
  9. Carver CS, & Harmon-Jones E (2009). Anger is an approach-related affect: evidence and implications. Psychological bulletin, 135(2), 183–204. 10.1037/a0013965 [DOI] [PubMed] [Google Scholar]
  10. Casaletto KB, & Heaton RK (2017). Neuropsychological Assessment: Past and Future. Journal of the International Neuropsychological Society : JINS, 23(9–10), 778–790. 10.1017/S1355617717001060 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Chin AA, Sweet AM, & Taylor CT (2023). Beyond Positive Affect: Discrete Positive Emotions Differentiate Major Depression from Social Anxiety Disorder. Cognitive Therapy and Research, 47(3), 377–385. 10.1007/s10608-023-10355-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Ciuk D, Troy AK, & Jones MC (2015). Measuring emotion: Self-reports vs. physiological indicators. SSRN Electronic Journal. 10.2139/ssrn.2595359 [DOI] [Google Scholar]
  13. Clark EA, Kessinger J, Duncan SE, Bell MA, Lahne J, Gallagher DL, & O’Keefe SF (2020). The Facial Action Coding System for Characterization of Human Affective Response to Consumer Product-Based Stimuli: A Systematic Review. Frontiers in psychology, 11, 920. 10.3389/fpsyg.2020.00920 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Coifman KG (forthcoming) Emotion and the Environment of Learning: Integrating Theories within a Developmental Framework. Affective Science for Clinical Scientists. Oxford University Press. [Google Scholar]
  15. Craft S, Newcomer J, Kanne S, Dagogo-Jack S, Cryer P, Sheline Y, Luby J, Dagogo-Jack A, &Alderson A (1996). Memory improvement following induced hyperinsulinemia in Alzheimer’s disease. Neurobiology of Aging, 17(1), 123–130. 10.1016/0197-4580(95)02002-0 [DOI] [PubMed] [Google Scholar]
  16. Cudo A, Francuz P, Augustynowicz P, & Stróżak P (2018). The Effects of Arousal and Approach Motivated Positive Affect on Cognitive Control. An ERP Study. Frontiers in human neuroscience, 12, 320. 10.3389/fnhum.2018.00320 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Dreisbach G (2006). How positive affect modulates cognitive control: the costs and benefits of reduced maintenance capability. Brain and cognition, 60(1), 11–19. 10.1016/j.bandc.2005.08.003 [DOI] [PubMed] [Google Scholar]
  18. Ekman P (1992). An argument for basic emotions. Cognition and Emotion, 6, 169–200. 10.1080/02699939208411068 [DOI] [Google Scholar]
  19. Ekman P, & Friesen WV (1978). Facial Action Coding System (FACS) [Database record]. APA PsycTests. 10.1037/t27734-000 [DOI] [Google Scholar]
  20. Elliot AJ, Eder AB, & Harmon-Jones E (2013). Approach–avoidance motivation and emotion: Convergence and divergence. Emotion Review, 5(3), 308–311. 10.1177/1754073913477517 [DOI] [Google Scholar]
  21. Fan Y, Lam JCK, & Li VOK (2021). Demographic effects on facial emotion expression: an interdisciplinary investigation of the facial action units of happiness. Scientific reports, 11(1), 5214. 10.1038/s41598-021-84632-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Field A (2020). Discovering statistics using IBM SPSS statistics. Sage. [Google Scholar]
  23. Gable P, & Harmon-Jones E (2010). The motivational dimensional model of affect: Implications for breadth of attention, memory, and cognitive categorisation. Cognition and Emotion, 24(2), 322–337. 10.1080/02699930903378305 [DOI] [Google Scholar]
  24. Galatzer-Levy IR, Abbas A, Koesmahargyo V, Yadav V, Perez-Rodriguez MM, Rosenfield P, Patil O, Dockendorf MF, Moyer M, Shipley LA, & Hansen BJ (2020). Facial and Vocal Markers of Schizophrenia Measured Using Remote Smartphone Assessments. 10.1101/2020.12.02.20219741 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Gao W, Yan X, & Yuan J (2022). Neural correlations between cognitive deficits and emotion regulation strategies: Understanding emotion dysregulation in depression from the perspective of cognitive control and cognitive biases. Psychoradiology, 2(3), 86–99. 10.1093/psyrad/kkac014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Garfinkel SN, Zorab E, Navaratnam N, Engels M, Mallorquí-Bagué N, Minati L, Dowell NG, Brosschot JF, Thayer JF, & Critchley HD (2016). Anger in brain and body: the neural and physiological perturbation of decision-making by emotion. Social cognitive and affective neuroscience, 11(1), 150–158. 10.1093/scan/nsv099 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Harmon-Jones C, Bastian B, & Harmon-Jones E (2016). The Discrete Emotions Questionnaire: A New Tool for Measuring State Self-Reported Emotions. PloS one, 11(8), e0159915. 10.1371/journal.pone.0159915 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Howard RS, Goldberg TE, Luo J, Munoz C, & Schneider LS (2023). Reliability of the NACC Telephone-administered Neuropsychological Battery (T-cog) and Montreal Cognitive Assessment for participants in the USC ADRC. Alzheimer’s & dementia, 15(1), e12406. 10.1002/dad2.12406 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Jung N, Wranke C, Hamburger K, & Knauff M (2014). How emotions affect logical reasoning: evidence from experiments with mood-manipulated participants, spider phobics, and people with exam anxiety. Frontiers in psychology, 5, 570. 10.3389/fpsyg.2014.00570 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kalanthroff E, Cohen N, & Henik A (2013). Stop feeling: inhibition of emotional interference following stop-signal trials. Frontiers in human neuroscience, 7, 78. 10.3389/fnhum.2013.00078 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Knowles KA, Cox RC, Armstrong T, & Olatunji BO (2019). Cognitive mechanisms of disgust in the development and maintenance of psychopathology: A qualitative review and synthesis. Clinical psychology review, 69, 30–50. 10.1016/j.cpr.2018.06.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Kreibig SD (2010). Autonomic nervous system activity in emotion: a review. Biological Psychology, 84(3), 394–421. 10.1016/j.biopsycho.2010.03.010 [DOI] [PubMed] [Google Scholar]
  33. Lai JM, Hawkins KA, Gross CP, & Karlawish JH (2008). Self-reported distress after cognitive testing in patients with Alzheimer’s disease. The journals of gerontology. Series A, Biological sciences and medical sciences, 63(8), 855–859. 10.1093/gerona/63.8.855 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Lezak MD, & Lezak MD (2004). Neuropsychological Assessment. Oxford University Press. [Google Scholar]
  35. Lichtenberger E, & Kaufman A (2013). Essentials of WAIS-IV Assessment, 2nd edition. New York: Wiley [Google Scholar]
  36. Lichtenberger E, & Kaufman A (2013). Essentials of WAIS-IV Assessment, 2nd edition. New York: Wiley [Google Scholar]
  37. Lindström BR, & Bohlin G (2012). Threat-relevance impairs executive functions: Negative impact on working memory and response inhibition. Emotion, 12(2), 384–393. 10.1037/a0027305 [DOI] [PubMed] [Google Scholar]
  38. McIntyre RS, Xiao HX, Syeda K, Vinberg M, Carvalho AF, Mansur RB, Maruschak N, & Cha DS (2015). The prevalence, measurement, and treatment of the cognitive dimension/domain in major depressive disorder. CNS drugs, 29(7), 577–589. 10.1007/s40263-015-0263-x [DOI] [PubMed] [Google Scholar]
  39. Meissner P, Poensgen C, & Wulf T (2021). How hot cognition can lead US astray: The effect of anger on strategic decision making. European Management Journal, 39(4), 434–444. 10.1016/j.emj.2020.09.010 [DOI] [Google Scholar]
  40. Millan MJ, Agid Y, Brüne M, Bullmore ET, Carter CS, Clayton NS, Connor R, Davis S, Deakin B, DeRubeis RJ, Dubois B, Geyer MA, Goodwin GM, Gorwood P, Jay TM, Joëls M, Mansuy IM, Meyer-Lindenberg A, Murphy D, Rolls E, Young LJ (2012). Cognitive dysfunction in psychiatric disorders: characteristics, causes and the quest for improved therapy. Nature reviews. Drug discovery, 11(2), 141–168. 10.1038/nrd3628 [DOI] [PubMed] [Google Scholar]
  41. Miller EK, & Cohen JD (2001). An integrative theory of prefrontal cortex function. Annual review of neuroscience, 24, 167–202. 10.1146/annurev.neuro.24.1.167 [DOI] [PubMed] [Google Scholar]
  42. Monsell SE, Dodge HH, Zhou XH, Bu Y, Besser LM, Mock C, Hawes SE, Kukull WA, Weintraub S, & Neuropsychology Work Group Advisory to the Clinical Task Force (2016). Results From the NACC Uniform Data Set Neuropsychological Battery Crosswalk Study. Alzheimer disease and associated disorders, 30(2), 134–139. 10.1097/WAD.0000000000000111 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Moons WG, & Shields GS (2015). Anxiety, not anger, induces inflammatory activity: An avoidance/approach model of immune system activation. Emotion, 15(4), 463–476. 10.1037/emo0000055 [DOI] [PubMed] [Google Scholar]
  44. Mrazik M, Millis S, & Drane DL (2010). The oral trail making test: effects of age and concurrent validity. Archives of Clinical Neuropsychology, 25(3), 236–243. 10.1093/arclin/acq006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Padmala S, Bauer A, & Pessoa L (2011). Negative emotion impairs conflict-driven executive control. Frontiers in Psychology, 2, 192. 10.3389/fpsyg.2011.00192 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Perini G, Cotta Ramusino M, Sinforiani E, Bernini S, Petrachi R, & Costa A (2019). Cognitive impairment in depression: recent advances and novel treatments. Neuropsychiatric disease and treatment, 15, 1249–1258. 10.2147/NDT.S199746 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Ricker JH, & Axelrod BN (1994). Analysis of an Oral Paradigm for the Trail Making Test. Assessment, 1(1), 47–52. 10.1177/1073191194001001007 [DOI] [PubMed] [Google Scholar]
  48. Rittenberg J (2017). Emotions in depression: What do we really know? Annual Review of Clinical Psychology, 13, 241–263. 10.1146/annurev-clinpsy-032816 [DOI] [PubMed] [Google Scholar]
  49. Roseman IJ (2013). Appraisal in the emotion system: Coherence in strategies for coping. Emotion Review, 5(2), 141–149. 10.1177/1754073912469591 [DOI] [Google Scholar]
  50. Russell JA (1991). Culture and the categorization of emotions. Psychological bulletin, 110(3), 426–450. 10.1037/0033-2909.110.3.426 [DOI] [PubMed] [Google Scholar]
  51. Sayette MA, Cohn JF, Wertz JM, Perrott MA, & Parrott DJ (2001). A psychometric evaluation of the facial action coding system for assessing spontaneous expression. Journal of Nonverbal Behavior, 25(3), 167–185. 10.1023/A:1010671109788 [DOI] [Google Scholar]
  52. Schroeder RW, Twumasi-Ankrah P, Baade LE, & Marshall PS (2012). Reliable Digit Span: a systematic review and cross-validation study. Assessment, 19(1), 21–30. 10.1177/1073191111428764 [DOI] [PubMed] [Google Scholar]
  53. Schultebraucks K, Yadav V, Shalev AY, Bonanno GA, & Galatzer-Levy IR (2022). Deep learning-based classification of posttraumatic stress disorder and depression following trauma utilizing visual and auditory markers of arousal and mood. Psychological medicine, 52(5), 957–967. 10.1017/S0033291720002718 [DOI] [PubMed] [Google Scholar]
  54. Shen L, Dillard JP, & Peng L (2022). Correspondence between two methods of measuring discrete emotions: Self-report versus machine-coded facial displays. Western Journal of Communication, 86(2), 194–214. 10.1080/10570314.2022.2040582 [DOI] [Google Scholar]
  55. Shields GS, Moons WG, Tewell CA, & Yonelinas AP (2016). The effect of negative affect on cognition: Anxiety, not anger, impairs executive function. Emotion, 16(6), 792–797. 10.1037/emo0000151 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Stern Y Cognitive reserve in ageing and Alzheimer’s disease. Lancet Neurol. 2012;11(11):1006–1012. doi: 10.1016/S1474-4422(12)70191-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Storbeck J, & Maswood R (2016). Happiness increases verbal and spatial working memory capacity where sadness does not: Emotion, working memory and executive control. Cognition & emotion, 30(5), 925–938. 10.1080/02699931.2015.1034091 [DOI] [PubMed] [Google Scholar]
  58. Threadgill AH, & Gable PA (2019). Negative affect varying in motivational intensity influences scope of memory. Cognition & emotion, 33(2), 332–345. 10.1080/02699931.2018.1451306 [DOI] [PubMed] [Google Scholar]
  59. Tyng CM, Amin HU, Saad MNM, & Malik AS (2017). The Influences of Emotion on Learning and Memory. Frontiers in psychology, 8, 1454. 10.3389/fpsyg.2017.01454 [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Wang Y, Chen J, & Yue Z (2017). Positive Emotion Facilitates Cognitive Flexibility: An fMRI Study. Frontiers in psychology, 8, 1832. 10.3389/fpsyg.2017.01832 [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Weintraub S, Besser L, Dodge HH, et al. (2018). Version 3 of the Alzheimer Disease Centers’ Neuropsychological Test Battery in the Uniform Data Set (UDS). Alzheimer Disease & Associated Disorders, 32, 10–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Wilson RS, Yu L, Lamar M, Schneider JA, Boyle PA, & Bennett DA (2019). Education and cognitive reserve in old age. Neurology, 92(10), e1041–e1050. https://doi-org.proxy.library.kent.edu/10.1212/WNL.0000000000007036 [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Yang H, Yang S, & Isen AM (2013). Positive affect improves working memory: implications for controlled cognitive processing. Cognition & emotion, 27(3), 474–482. 10.1080/02699931.2012.713325 [DOI] [PubMed] [Google Scholar]
  64. Yerkes RM & Dodson JD (1908). The relationship of strength of stimulus to rapidity of habit formation. Journal of Comparative Neurology and Psychology, 18, 459–482. [Google Scholar]

RESOURCES