Login
Facebook YouTube Twitter Linkedin
Facebook YouTube Twitter Linkedin

Blog

7 November 2018

Emotion recognition biases in alcohol use disorder

Researcher at the National Institute on Alcohol Abuse and Alcoholism (NIAAA), Clara Freeman, shared with us about why she chose the CANTAB Emotion Recognition Task for her latest publication

Can you tell us more about your research group?

The Laboratory of Neuroimaging, led by Dr Nora Volkow, is part of the National Institute on Alcohol Abuse and Alcoholism (NIAAA). Our lab’s mission is to understand the neurocircuitry of addiction and the rewarding effects of drugs and natural reinforcers in the human brain. While our primary methods include PET and MRI, we are also interested in neuropsychological deficits associated with addictive behavior.

 

What is the rationale behind your study?

Previous studies have indicated that alcohol use disorder (AUD) is associated with deficits in the recognition of certain emotional facial expressions (Castellano et al., 2015). However, results regarding the specific manifestations of these deficits were mixed. Furthermore, it was unclear whether individuals with AUD were worse at identifying facial expressions because of something specific to alcohol use or because of accompanying depressed mood or anxiety. In this study, we tried to clarify the nature of these deficits and determine whether they were unique to AUD, or instead related to depressed mood or anxiety.  

 

Which methods did you use?

We assessed facial emotion recognition ability using CANTAB’s Emotion Recognition Task (ERT). Our variables of interest included latency and accuracy in identifying the six core emotions, happiness, surprise, sadness, anger, fear, and disgust, at varying levels of intensity. We also assessed misidentification of emotions, a measure of bias toward a given emotion, calculated by dividing incorrect selections by total selections of that emotion.

We compared task performance between a sample of 19 AUD patients and 19 healthy controls. We assessed drinking behavior from the past 90 days using the Timeline Followback interview (Sobell and Sobell, 2000), depressed mood using the Beck Depression Inventory (Beck et al., 1998) and trait anxiety using the Spielberger Trait Anxiety Inventory (Spielberger et al., 1983).  

 

What are your key findings?

The study showed that there were no differences in latency or accuracy in emotion identification between our AUD and control groups; however, there were group differences in their pattern of errors. While the control group tended to misattribute happiness to ambiguous faces, the AUD group showed a bias toward anger and disgust. In the AUD participants, these biases were positively correlated with the number of drinks consumed in the last 90 days, but was not associated with depression or anxiety scores.

 

What are the implications of your study?

This is the first study to show that emotion recognition biases in alcohol use disorder worsen with increased drinking quantity and are not due to associated mood symptoms. Our results suggest that individuals with alcohol use disorder may interpret neutral or nonthreatening facial expressions as hostile. A hostile attribution bias has been associated with aggression and is maladaptive for forming and maintaining positive relationships. Interventions designed to attenuate this hostility bias may improve treatment results and interpersonal functioning for patients with an AUD.

 

Why did you choose CANTAB?

Previous research on AUD and facial emotion recognition has used various tasks, which may contribute to inconsistency in results. Using a validated and accessible task like CANTAB allows for easier standardization and replication of the research. Another key advantage of CANTAB’s Emotion Recognition Task is that it includes multiple variables beyond mere accuracy, allowing for a richer analysis.   

 

What are the next steps for your research?

We are interested in understanding the mechanisms behind these emotion recognition biases. Our primary future direction involves using the ERT in an fMRI study design. This way we can evaluate if abnormal patterns of functional connectivity in brain regions critical for emotion processing underlie the group differences between individuals with AUD and healthy controls. We are also interested in investigating whether these biases predict clinical outcomes and decreased social relationships in the real world.

 

Interested in assessing emotion recognition?

Contact us 

 

References

Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Beck depression inventory-II. San Antonio, 78(2), 490-498.

Castellano, F., Bartoli, F., Crocamo, C., Gamba, G., Tremolada, M., Santambrogio, J., Clerici, M. and Carrà, G. (2015). Facial emotion recognition in alcohol and substance use disorders: a meta-analysis. Neuroscience & Biobehavioral Reviews, 59, pp.147-154.

Freeman, C. R., Wiers, C. E., Sloan, M. E., Zehra, A., Ramirez, V., Wang, G.-J., & Volkow, N. D. (2018). Emotion Recognition Biases in Alcohol Use Disorder. Alcoholism: Clinical and Experimental Research, 42(8), 1541–1547. http://doi.org/10.1111/acer.1380

Sobell, L.C. & Sobell, M.B. (2000). Alcohol Timeline Followback (TFLB). In American Psychiatric Association (Ed.), Handbook of Psychiatric Measures (pp. 477-479). Washington, DC: American Psychiatric Association.

Spielberger, C.D., Gorsuch, R.L., Lushene, R.E. and Vagg, P.R., (2010). State-trait anxiety inventory (STAI). 1970, p.180.

Tags : emotion recognition | emotion bias

Author portrait

Clara Freeman, NIAAA Researcher