Emotion recognition of morphed facial expressions in presymptomatic and symptomatic frontotemporal dementia, and Alzheimer’s disease

Abstract

AbstractBackgroundAn increasing amount of studies are emphasizing the importance of social cognitive assessments to improve early diagnosis between frontotemporal dementia (FTD) and Alzheimer’s Disease (AD). Traditional measures of social cognition, including emotion recognition paradigms, however seem insensitive to the subtle deficits in the earliest dementia stages. The Emotion Recognition Task (ERT) was developed to identify early deficits in the recognition of facial expressions across different levels of intensity.MethodUsing the ERT, we assessed the recognition of morphed facial expressions of six emotions (anger, disgust, fear, happiness, sadness and surprise) across four intensities (40, 60, 80 and 100%) in patients with behavioural variant FTD (bvFTD; n=32), patients with AD (n=32), FTD mutation carriers (presymptomatic n=47, mildly/questionably symptomatic n=5) and controls (n=49). We examined group differences using multilevel linear regression modelling, and performed post hoc analyses on presymptomatic (MAPT, GRN and C9orf72) and mildly/questionably symptomatic mutation carriers. Classification abilities were investigated with logistic regression analyses; we determined sensitivity/specificity with ROC analyses.ResultThe lowest ERT total scores were found in bvFTD and AD patients, whereas equally highest scores were found in mutation carriers and controls. bvFTD patients scored significantly lower than AD patients on anger (p=0.023) and happiness (p<0.001), where subjects with mild/questionable FTD had an intermediate position between patients with dementia and those without (p=0.051) (Figure). Discriminative ability was better at the lowest (40%) than at the highest intensity (100%). Presymptomatic C9orf72 mutation carriers were worse at recognizing anger at the lowest (40%) emotional intensity than controls (p=0.038) and presymptomatic GRN mutation carriers (p=0.047). The ERT classified between patients with bvFTD and controls, patients with AD and controls (both p<0.001), but not between patients with bvFTD and AD, nor the total group of mutation carriers and controls (p>0.05).ConclusionOur results demonstrate clear emotion recognition deficits in bvFTD and AD, and suggest the presence of subtle changes in facial emotion recognition in early‐stage FTD. This provides us more insight into the discrete neural substrates involved in behavioural and emotional changes in developing FTD, but also highlights the importance of incorporating emotion recognition paradigms into standard neuropsychological assessment for early differential diagnosis.