The female advantage: sex as a possible protective factor against emotion recognition impairment following traumatic brain injury

Social impairment is common in individuals with moderate to severe traumatic brain injury (TBI; Andrews, Rose, & Johnson, 1998; Duff, Mutlu, Byom, & Turkstra, 2012; Gomez-Hernandez, Max, Kosier, Paradiso, & Robinson, 1997; Milders, Fuchs, & Crawford, 2003; Temkin, Corrigan, Dikmen, & Machamer, 2009; Ylvisaker, Turkstra, & Coelho, 2005) and is a major predictor of overall outcome (Morton & Wehman, 1995). Facial affect recognition ability, in particular, is strongly associated with negative social outcomes, such as poor social integration (Knox & Douglas, 2009) and reports of socially inappropriate behaviors (Pettersen, 1991).

A recent meta-analysis estimated that up to 39% of individuals in the chronic phase of a moderate to severe TBI are significantly impaired at recognizing facial affect (Babbage et al., 2011). To date, however, there are no methods for identifying which patients are at risk for emotion recognition deficits. Rosenberg, Dethier, Kessels, Westbrook, and McDonald, (2015) provided some evidence that TBI severity (measured by posttraumatic amnesia) is a risk factor (Rosenberg et al., 2015), but severity accounted for only 37% of the variance in affect-recognition task scores. Another candidate predictor of impairments is site of brain damage, because TBI often affects brain regions implicated in affect recognition. However, Green, Turner, and Thompson (2004) reported that participants with TBI who had focal damage in regions involved in emotion recognition—including right posterior hemisphere, amygdala, basal ganglia—did not perform significantly worse than participants with TBI who had damage in other brain regions (Green et al., 2004). It has been hypothesized that the degree of emotion recognition impairment might correlate with the degree of frontal or fronto-temporal functioning (Bornhofen & McDonald, 2008; Hornak, Rolls, & Wade, 1996; Olson, McCoy, Klobusicky, & Ross, 2013; Willis, Murphy, Ridley, & Vercammen, 2015). While this claim has some support from the TBI literature (Martins et al., 2012; Spikman, Timmerman, Milders, Veenstra, & van der Naalt, 2012), findings to date remain inconclusive.

There is some evidence that facial affect recognition abilities are related to other nonsocial cognitive skills, such as verbal and nonverbal working memory and processing speed (Yim, Babbage, Zupan, Neumann, & Willer, 2013), verbal concept formation and alternating fluency (Williams & Wood, 2010), and cognitive flexibility (Milders et al., 2003). Other studies, however, have found no correlations between emotion recognition and these skills (Rosenberg et al., 2015; Spikman et al., 2012).

A final potential predictor of emotion recognition impairments is sex of the participant. There is a long history of research showing a female advantage in recognizing emotions (Kret & De Gelder, 2012; Montagne, Kessels, Frigerio, de Haan, & Perrett, 2005; Weisenbach et al., 2014) beginning in infancy and continuing throughout the life span (Kessels, Montagne, Hendriks, Perrett, & de Haan, 2014; McClure, 2000). Thus, women with TBI might have an advantage that would protect them from affect recognition impairments, whereas men with TBI might be more vulnerable. To date, examinations of the relationship between sex and TBI outcome, however, have yielded mixed results. For instance, data from the IMPACT study showed that sex is not a significant predictor of 6-month overall outcome as measured by the Glasgow Outcome Scale (Mushkudiani et al., 2007), but a previous meta-analysis by Farace and Alves (2000) revealed that women with TBI had worse outcome than men. These mixed results are likely to be due to the multiplicity of measures that have been used to assess outcome. Indeed, breaking outcome down into the cognitive domain highlights the complexity of the relationship between cognitive abilities and sex following TBI; women outperform men in language, attention- and working-memory tasks (Ratcliff et al., 2007), and visual memory (Moore, Ashman, Cantor, Krinick, & Spielman, 2010), whereas the opposite is true for visual analytic skills (Ratcliff et al., 2007).

Recent work on TBI populations has reported sex-based differences in the social domain. Despins, Turkstra, Struchen, and Clark (2015) found that women with TBI rated their pragmatic communication skills more accurately than did men with comparable injury severity. Adult females with TBI are better than males at theory-of-mind tasks (Turkstra, 2008), and they show higher rates of community integration years after their injury (R. L. Wood & Rutterford, 2006). A longitudinal study on pediatric TBI found that girls not only performed better on a facial affect-recognition task at baseline but also improved their skills at a higher rate than boys (Schmidt, Hanten, Li, Orsten, & Levin, 2010). These findings, coupled with clinical evidence that women experience better social outcomes than men after TBI (Farace & Alves, 2000), support the notion that women may have better resilience to TBI-related social deficits. However, there is little experimental evidence for an interaction between sex and brain injury in emotion recognition abilities. Further clarification of the role of sex in social functioning in general, and in facial affect recognition in particular, is crucial; not only does such understanding have the potential to inform clinicians of subgroups of patients with TBI who are more likely to develop emotion recognition impairment and may benefit from early rehabilitation, but it can also inform the development of new behavioral and pharmacological treatments tailored to men or women.

In the current study, we sought to examine the relationship between sex and the ability to label facial affect in adults with TBI. Based on evidence of sex-differences in social functioning in both TBI and healthy populations, we hypothesized that being female would serve as a protective factor for emotion recognition impairment following TBI. To test this hypothesis, we administered two different emotion-labeling tasks to a sample of individuals with moderate to severe TBI and a demographically matched comparison group. The first task was a traditional emotion recognition test, in which participants were asked to label emotions displayed by actors in photographs. The second was a dynamic facial affect-recognition task, in which stimuli included morphed images that presented a range of intensities of emotion expression. Dynamic tasks including different intensities have been argued to be more ecologically valid and to reveal individual differences within healthy populations (Montagne, Kessels, De Haan, & Perrett, 2007; Wehrle, Kaiser, Schmidt, & Scherer, 2000). Our main predictions were a main effect of group, resulting in higher accuracy on both tasks in the comparison group, and an interaction of group by sex, manifested as higher overall scores in women and a larger sex-based difference between women and men with TBI. Moreover, we investigated the presence of a three-way interaction among group, sex, and emotion type in order to explore whether the sex-based difference was specific to individual emotions. Last, we aimed to determine how sex influences the relationship between emotion recognition abilities and performance on tasks measuring “nonsocial” cognitive skills, such as executive functioning, verbal learning and memory, processing speed, and self-reported psychological distress.

DOI: 10.3758/s13415-016-0437-0

BibTex

@article{Rigon_2016,
	doi = {10.3758/s13415-016-0437-0},
	url = {https://doi.org/10.3758%2Fs13415-016-0437-0},
	year = 2016,
	month = {may},
	publisher = {Springer Science and Business Media {LLC}},
	volume = {16},
	number = {5},
	pages = {866--875},
	author = {Arianna Rigon and Lyn Turkstra and Bilge Mutlu and Melissa Duff},
	title = {The female advantage: sex as a possible protective factor against emotion recognition impairment following traumatic brain injury},
	journal = {Cognitive, Affective, {\&} Behavioral Neuroscience}
}

@article{Rigon_2016,
	doi = {10.3758/s13415-016-0437-0},
	url = {https://doi.org/10.3758%2Fs13415-016-0437-0},
	year = 2016,
	month = {may},
	publisher = {Springer Science and Business Media {LLC}},
	volume = {16},
	number = {5},
	pages = {866--875},
	author = {Arianna Rigon and Lyn Turkstra and Bilge Mutlu and Melissa Duff},
	title = {The female advantage: sex as a possible protective factor against emotion recognition impairment following traumatic brain injury},
	journal = {Cognitive, Affective, {\&} Behavioral Neuroscience}
}