Emotion recognition after traumatic brain injury (TBI)
Research type
Research Study
Full title
Investigating emotion recognition after traumatic brain injury (TBI) using the Bristol Emotion Recognition Task (ERT) and exploring the potential impact of influencing factors.
IRAS ID
219738
Contact name
Maren Mueller-Glodde
Contact email
Sponsor organisation
University of Bristol
Clinicaltrials.gov Identifier
https://osf.io/exs39/, Open Science Network; R&I# 4238, Research & Innovation at NBT
Duration of Study in the UK
1 years, 0 months, 10 days
Research summary
Research Summary
The study is investigating emotion recognition in people who have suffered traumatic brain injury (TBI). Previous research suggests that people with TBI have difficulties recognising emotions from facial expressions (e.g. Babbage et al. 2011), which in turn could impact their ability to interact successfully with other people (e.g. Knapp, Hall, & Horgan, 2013). However, emotion recognition is also impacted by other factors including anxiety (e.g. Attwood et al., 2017), which is prevalent after TBI. To appropriately address difficulties that patients are presenting with it is vital to understand potential underlying associations.\nThis study is using the Bristol Emotion Recognition Task (ERT) to investigate potential deficits and biases in emotion recognition after TBI, as well as investigating the influence of other factors, such as anxiety, depression, alexithymia, aggression, and cognitive ability. The impact of TBI on emotion recognition will be assessed in patients with TBI and in a non-brain injury control group consisting of friends and relatives of these patients. Following that the impact of other factors, in particular anxiety, will be explored to further our understanding of emotion recognition after TBI.\nPatients who have TBI and their family members or friends will be recruited through the Neuropsychology department at the North Bristol NHS Trust and the Frenchay Head Injury Therapy Unit. After giving consent participants will be asked to complete some questionnaires at home and then attend a one-hour testing session during which they will be asked to complete a series of computer-based tasks and a further questionnaire. At the end of that session they will be asked to confirm that they are happy for the data to be used in study and their participation will end.
Summary of Results
Facial expressions are an important source of information about other people’s state of mind and emotions and are commonly used to guide our social interactions. Research suggests that emotion recognition is generally lower for people with moderate to severe Traumatic Brain Injury (Babbage et al., 2011; Murphy et al., 2021). Although we know this, emotion recognition is not commonly assessed in a clinical setting because a lot of tasks used to measure emotion recognition have not been fully evaluated (Kelly et al., 2017). The Bristol Emotion Recognition Task (Bristol ERT) is an emotion recognition task available as part of the Cambridge Automated Neuropsychological Test Battery: https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fu2790089.ct.sendgrid.net%2Fls%2Fclick%3Fupn%3DXv3JSvJ-2B3M71ppf7N9agbUdV2cpGD6KaqDfW9i4jg-2BNRZOcgxnhc5X8DJbs64yT-2FIW78lK9CArrPA7-2B1TzfiGhaZmg5mDZW7bYZnHxyRXDVImGGY6l1hjqWDetOYRvK8rlDMJhtnOqgmDU7P52o-2F4G7Gn4QstJ7scSAz-2FCSDstI-3DOuii_E1aO2-2BZlVOSJJV-2FajQqskegTd6IRomHYTi-2Fbt8SH3YIZL3qD67KTUpaYb6h7jC1KGWJdyk-2BCxJTkSD6w0RSm3sp0dIFA4M2Zd41E7N2wKbHMVbPRlxagxxYtE-2FdnBviq1j60YGLaRY2vmUrSK3EvmVKTiY6jL6mzLLpSoKL0mb-2FL2CxUbHkcFV4IZiFrxe4lChrRmC3-2FPrAa6WCBE2jayQ-3D-3D&data=05%7C01%7Capprovals%40hra.nhs.uk%7Cdd7b5ae1227b4569a3b008da69be2e88%7C8e1f0acad87d4f20939e36243d574267%7C0%7C0%7C637938562834757699%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=JZjPZT6HIGQAojJhiOLbl%2FMpM9aCGthoEiZfZE0qkCI%3D&reserved=0
The main aim of this study was to see whether the Bristol ERT was able to detect a difference in emotion recognition between people who have had a moderate to severe Traumatic Brain Injury and people without a brain injury. There are many factors that can influence how people perform on emotion recognition tasks, so we also included so called potentially confounding factors in this study. This helped us understand whether the Bristol ERT could be useful in a clinical setting.
In line with previous studies (Babbage et al., 2011), the results from this study suggested that having moderate to severe Traumatic Brain Injury is associated with decreased emotion recognition. The size of the effect observed in this study was large (Hedge’s g = 1.28) and is similar to the effect size reported by Babbage and colleagues (2011) when they combined data from different studies. The potentially confounding factors included in this study were measures of anxiety, face perception, alexithymia (ability to recognise and describe one’s own emotions), depression, stress, and aggression. The results suggested that the lower emotion recognition scores in the Traumatic Brain Injury group compared to controls were not explained by any of the confounding factors included. We also wanted to know whether people with Traumatic Brain Injury were more likely to identify ambiguous facial expressions as negative. There was no evidence that having a moderate to severe TBI was associated with bias on either of the two Bristol EBT suggesting that participants with TBI did not show a negative attribution bias in this study. Overall, the findings in this study show that the Bristol ERT is suitable for use in a moderate to severe TBI population and could be a useful task to use in a clinical setting.
An important part of evaluating tasks for clinical use is understanding whether reliable measurements can be made using that task. Internal consistency reliability considers whether response made are consistent across trials on the Bristol ERT. The reliability estimates for the Bristol ERT in this study were high based on the descriptors used by Strauss and colleagues (2006), which means that measurements made were likely reliable. Notably, the estimates were above 0.8, which has been suggested as a minimally acceptable threshold of reliability for tasks used in individual assessment (Ponterotto & Ruckdeschel, 2007; Sattler, 2001). However, it is important to consider that the Smallest Real Difference calculated using the reliability estimates suggests a fairly big change in individual performance is needed to be confident in a true change plus or minus 13 correct responses out of a possible 96). Further studies are needed to evaluate the Bristol ERT in a Traumatic Brain Injury population before the task can be used clinically, the findings from this study are promising. It is also important to consider whether the decreased performance on the Bristol ERT observed after moderate to severe Traumatic Brain Injury corresponds to a noticeable change in functioning or behaviour.
REC name
South West - Central Bristol Research Ethics Committee
REC reference
19/SW/0062
Date of REC Opinion
30 May 2019
REC opinion
Further Information Favourable Opinion