Facial Emotion Recognition Performance Research Articles

Attention-Guided Network Model for Image-Based Emotion Recognition

Neural networks are increasingly able to outperform traditional machine learning and filtering approaches in classification tasks. However, with the rise in their popularity, many unknowns still exist when it comes to the internal learning processes of the networks in terms of how they make the right decisions for prediction. As a result, in this work, different attention modules integrated into a convolutional neural network coupled with an attention-guided strategy were examined for facial emotion recognition performance. A custom attention block, AGFER, was developed and evaluated against two other well-known modules of squeeze–excitation and convolution block attention modules and compared with the base model architecture. All models were trained and validated using a subset from the OULU-CASIA database. Afterward, cross-database testing was performed using the FACES dataset to assess the generalization capability of the trained models. The results showed that the proposed attention module with the guidance strategy showed better performance than the base architecture while maintaining similar results versus other popular attention modules. The developed AGFER attention-integrated model focused on relevant features for facial emotion recognition, highlighting the efficacy of guiding the model during the integral training process.

Network dysfunction of sadness facial expression processing and morphometry in euthymic bipolar disorder

Facial emotion recognition (FER), including sadness, is altered in bipolar disorder (BD). However, the relationship between this impairment and the brain structure in BD is relatively unexplored. Furthermore, its association with clinical variables and with the subtypes of BD remains to be clarified. Twenty euthymic patients with BD type I (BD-I), 28 BD type II (BD-II), and 45 healthy controls completed a FER test and a 3D-T1-weighted magnetic resonance imaging. Gray matter volume (GMV) of the cortico-limbic regions implicated in emotional processing was estimated and their relationship with FER performance was investigated using network analysis. Patients with BD-I had worse total and sadness-related FER performance relative to the other groups. Total FER performance was significantly negatively associated with illness duration and positively associated with global functioning in patients with BD-I. Sadness-related FER performance was also significantly negatively associated with the number of previous manic episodes. Network analysis showed a reduced association of the GMV of the frontal–insular–occipital areas in patients with BD-I, with a greater edge strength between sadness-related FER performance and amygdala GMV relative to controls. Our results suggest that FER performance, particularly for facial sadness, may be distinctively impaired in patients with BD-I. The pattern of reduced interrelationship in the frontal–insular–occipital regions and a stronger positive relationship between facial sadness recognition and the amygdala GMV in BD may reflect altered cortical modulation of limbic structures that ultimately predisposes to emotional dysregulation. Future longitudinal studies investigating the effect of mood state on FER performance in BD are warranted.

Facial emotion recognition using deep learning detector and classifier

Numerous research works have been put forward over the years to advance the field of facial expression recognition which until today, is still considered a challenging task. The selection of image color space and the use of facial alignment as preprocessing steps may collectively pose a significant impact on the accuracy and computational cost of facial emotion recognition, which is crucial to optimize the speed-accuracy trade-off. This paper proposed a deep learning-based facial emotion recognition pipeline that can be used to predict the emotion of detected face regions in video sequences. Five well-known state-of-the-art convolutional neural network architectures are used for training the emotion classifier to identify the network architecture which gives the best speed-accuracy trade-off. Two distinct facial emotion training datasets are prepared to investigate the effect of image color space and facial alignment on the performance of facial emotion recognition. Experimental results show that training a facial expression recognition model with grayscale-aligned facial images is preferable as it offers better recognition rates with lower detection latency. The lightweight MobileNet_v1 is identified as the best-performing model with WM=0.75 and RM=160 as its hyper-parameters, achieving an overall accuracy of 86.42% on the testing video dataset.

FERFM: An Enhanced Facial Emotion Recognition System Using Fine-tuned MobileNetV2 Architecture

Deep learning techniques have recently drawn a lot of attention towards automatic facial emotion recognition (FER) applications such as diagnosis of psychiatric disorders, mental health monitoring, and other human–computer interaction applications. But the models that have been developed, have still not been generalized due to the absence of substantial emotion datasets and high operational costs. To address these issues, we propose a novel technique called facial emotion recognition using fine-tuned MobileNetV2 architecture (FERFM). It uses the concept of transfer learning to improve the performance of FER systems for mobile devices. It transfers the knowledge from pre-trained ImageNet dataset to profile view RGB-KDEF dataset for diversity among learnt features. Initially, data descriptive analysis is performed on input KDEF images to automate the necessary knowledge. Then the pipeline strategy is introduced, where the pre-trained MobileNetV2 architecture is fine tuned by eliminating the last six layers and adding dropout regularization, max pooling and dense layer. To test the performance of the proposed work, we ran several tests using the KDEF-RGB dataset, FER13 dataset and real-time facial expression images to classify the results into seven discrete categories of emotions. It was also verified on VGG-16 pretrained model and demonstrated remarkable performance comparatively. Proposed FERFM achieves 85.7% accuracy and requires 1,510,599 number of trainable parameters, 43 ms running time per image that proves the enhancement of FER performance and outperforms the existing state of art methods. Also, the ability to achieve low operational costs while maintaining high accuracy demonstrates proficiency in mobile applications.

Facial Emotion Recognition in Older Adults With Cognitive Complaints.

Facial emotion recognition deficits impact the daily life, particularly of Alzheimer's disease patients. We aimed to assess these deficits in the following three groups: subjective cognitive decline (SCD), mild cognitive impairment (MCI), and mild Alzheimer's dementia (AD). Additionally, we explored the associations between facial emotion recognition and cognitive performance. We used the Korean version of the Florida Facial Affect Battery (K-FAB) in 72 SCD, 76 MCI, and 76 mild AD subjects. The comparison was conducted using the analysis of covariance (ANCOVA), with adjustments being made for age and sex. The Mini-Mental State Examination (MMSE) was utilized to gauge the overall cognitive status, while the Seoul Neuropsychological Screening Battery (SNSB) was employed to evaluate the performance in the following five cognitive domains: attention, language, visuospatial abilities, memory, and frontal executive functions. The ANCOVA results showed significant differences in K-FAB subtests 3, 4, and 5 (p=0.001, p=0.003, and p=0.004, respectively), especially for anger and fearful emotions. Recognition of 'anger' in the FAB subtest 5 declined from SCD to MCI to mild AD. Correlations were observed with age and education, and after controlling for these factors, MMSE and frontal executive function were associated with FAB tests, particularly in the FAB subtest 5 (r=0.507, p<0.001 and r=-0.288, p=0.026, respectively). Emotion recognition deficits worsened from SCD to MCI to mild AD, especially for negative emotions. Complex tasks, such as matching, selection, and naming, showed greater deficits, with a connection to cognitive impairment, especially frontal executive dysfunction.

Impact of facemasks on psychotherapy: Clinician's confidence and emotion recognition.

Facial emotion recognition is a key component of human interactions, and in clinical relationships contributes to building and maintaining the therapeutic alliance with patients. The introduction of facemasks has reduced the availability of facial information in private and professional relationships. This study aimed to assess the impact of facemasks on clinicians' perception of clinical interactions as well as their ability to read facial expressions. In this cross-sectional study, a purposive sample of 342 clinical psychologists or psychotherapists completed an online survey including the assessment of burnout, alexithymia, emotion dysregulation, and self-perceived ability to build effective relationships and communication with patients with/without facemasks. Participants were randomly assigned to the standardized facial emotion recognition task Diagnostic Analysis of Nonverbal Accuracy FACES 2-Adult Faces including 24 faces representing anger, fear, sadness, and happiness. Facemasks impaired the self-perceived ability of clinicians to build effective relationships and communicate with patients and reduced satisfaction in clinical encounters. The ability of clinicians to recognize facial emotions is significantly reduced for masked happy and angry faces, but not for sad and afraid ones. The perceived difficulty in building good relationships and communication with patients had a positive correlation with alexithymia and emotion dysregulation; higher levels of discomfort when wearing facemasks had a positive correlation with burnout and emotion dysregulation. Facemasks reduced clinicians' self-confidence in clinical encounters with patients wearing facemasks, but their facial emotion recognition performance was only partially impaired.

Difficulties in facial emotion recognition: taking psychopathic and alexithymic traits into account

BackgroundAlexithymic and psychopathic traits are closely associated, but distinct constructs. Both have been associated with facial emotion processing deficits reflecting empathy deficits, however the underlying mechanism contributing to these deficits is not well-understood.MethodsThis study investigated facial emotion recognition performance in a non-clinical sample (N = 110) including both male and female participants, with varying levels of psychopathic and alexithymic traits. Facial emotion recognition was assessed using a computerized task, that presented different expressions depicting five emotional states (pain, fear, sadness, anger, happiness) and neutral expressions.ResultsResults suggest that the high psychopathic traits group reported lower accuracy compared to the low psychopathic traits group, indicating a more generalized deficit in facial affect recognition, across all emotions. The alexithymic groups (high vs. low) on the other hand did not differ in their performance on the task for any of the emotions presented.ConclusionThese findings add to the current body of research regarding face processing categorization deficits in relation to psychopathic and alexithymic traits and can inform prevention and intervention efforts that aim to facilitate facial emotion recognition in individuals with these personality traits.

A Single Session of Bifrontal tDCS Can Improve Facial Emotion Recognition in Major Depressive Disorder: An Exploratory Pilot Study.

Emotional processing deficits are key features in major depressive disorder (MDD). Neuroimaging studies indicate that the dorsolateral prefrontal cortex (DLPFC) plays a pivotal role in both depressive symptoms and emotional processing. Recently, transcranial Direct Current Stimulations (tDCS) applied over the DLPFCs have held the promise to alleviate the symptoms in patients with MDD, but the effect on emotional processing in the patients is unclear. Here, we investigated the effect of a single session of tDCS over the DLPFCs on the emotional processing in patients with treatment-resistant MDD. In a randomized sham-controlled study, 35 patients received a single 30 min session of either active (2 mA, n = 18) or sham tDCS (n = 17). The anode was placed over the left and the cathode over the right DLPFC. Emotional processing accuracy was measured by a facial emotion recognition (FER) task. We observed an overall improvement in FER performance after the active tDCS, but not the sham tDCS. These exploratory results suggest that a single session of tDCS over the DLPFCs may improve FER in MDD, a crucial function of social cognition. Further studies are needed to investigate whether this acute improvement of FER in response to a single tDCS session could translate into clinical benefits or predict remission following repeated sessions of stimulation.

Facial emotion recognition in children and youth with attention-deficit/hyperactivity disorder and irritability.

The ability to recognize emotions evident in people's faces contributes to social functioning and might be affected by ADHD and irritability. Given their high co-occurrence, we examined the relative contribution of ADHD and irritability to facial emotion recognition (FER). We hypothesized that irritability but not ADHD traits would predict increased likelihood of misrecognizing emotions as negative, and that FER performance would explain the association of ADHD and irritability traits with social skills. FER was measured using the Reading the Mind in the Eyes Test (RMET) in children (6-14years old) referred for ADHD assessment (n = 304) and healthy controls (n = 128). ADHD, irritability and social skills were measured using parent ratings. We used repeated measure logistics regression, comparing the effects across emotion valence of images (i.e., neutral/positive/negative). High irritability but not ADHD diagnosis predicted lower RMET accuracy. ADHD traits predicted lower RMET accuracy in younger but not older participants, whereas irritability predicted poorer accuracy at all ages. ADHD traits predicted lower RMET accuracy across all emotion valences, whereas irritability predicted increased probability of misrecognizing neutral and positive but not negative emotions. Irritability did not increase the probability for erroneously recognizing emotions as negative. ADHD and irritability traits fully explained the association between RMET and social skills. ADHD and irritability traits might impact the ability to identify emotions portrayed in faces. However, irritability traits appear to selectively impair recognition of neutral and positive but not negative emotions. ADHD and irritability are important when examining the link between FER and social difficulties.

The association between cannabis use and facial emotion recognition in schizophrenia, siblings, and healthy controls: Results from the EUGEI study

Schizophrenia is frequently accompanied with social cognitive disturbances. Cannabis represents one established environmental factor associated with the onset and progression of schizophrenia. The present cross-sectional study aimed to investigate the association of facial emotion recognition (FER) performance with cannabis use in 2039 patients with schizophrenia, 2141 siblings, and 2049 healthy controls (HC). FER performance was measured using the Degraded Facial Affect Recognition Task (DFAR). Better FER performance as indicated by higher DFAR-total scores was associated with lifetime regular cannabis use in schizophrenia (B = 1.36, 95% CI 0.02 to 2.69), siblings (B = 2.17, 95% CI 0.79 to 3.56), and HC (B = 3.10, 95% CI 1.14 to 5.06). No associations were found between DFAR-total and current cannabis use. Patients with schizophrenia who started to use cannabis after the age of 16 showed better FER performance than patients who started earlier (B = 2.50, 95% CI 0.15 to 4.84) and non-users (B = 3.72, 95 CI 1.96 to 5.49). Better FER performance was found also in siblings who started to use cannabis after 16 compared to non-users (B = 2.37, 95% CI 0.58 to 4.16), while HC using cannabis performed better than non-users at DFAR-total regardless of the age at onset. Our findings suggest that lifetime regular cannabis use may be associated with better FER regardless of the psychosis risk, but that FER might be moderated by age at first use in people with higher genetic risk. Longitudinal studies may clarify whether there is a cause-and-effect relationship between cannabis use and FER performance in psychotic and non-psychotic samples.

Efficient facial emotion recognition model using deep convolutional neural network and modified joint trilateral filter

Facial emotion recognition extracts the human emotions from the images and videos. As such, it requires an algorithm to understand and model the relationships between faces and facial expressions and to recognize human emotions. Recently, deep learning models are utilized to improve the performance of facial emotion recognition. However, the deep learning models suffer from the overfitting issue. Moreover, deep learning models perform poorly for images which have poor visibility and noise. Therefore, in this paper, an efficient deep learning-based facial emotion recognition model is proposed. Initially, contrast-limited adaptive histogram equalization (CLAHE) is applied to improve the visibility of input images. Thereafter, a modified joint trilateral filter is applied to the obtained enhanced images to remove the impact of impulsive noise. Finally, an efficient deep convolutional neural network is designed. Adam optimizer is also utilized to optimize the cost function of deep convolutional neural networks. Experiments are conducted by using the benchmark dataset and competitive human emotion recognition models. Comparative analysis demonstrates that the proposed facial emotion recognition model performs considerably better compared to the competitive models

Foveal processing of emotion-informative facial features.

Certain facial features provide useful information for recognition of facial expressions. In two experiments, we investigated whether foveating informative features of briefly presented expressions improves recognition accuracy and whether these features are targeted reflexively when not foveated. Angry, fearful, surprised, and sad or disgusted expressions were presented briefly at locations which would ensure foveation of specific features. Foveating the mouth of fearful, surprised and disgusted expressions improved emotion recognition compared to foveating an eye or cheek or the central brow. Foveating the brow led to equivocal results in anger recognition across the two experiments, which might be due to the different combination of emotions used. There was no consistent evidence suggesting that reflexive first saccades targeted emotion-relevant features; instead, they targeted the closest feature to initial fixation. In a third experiment, angry, fearful, surprised and disgusted expressions were presented for 5 seconds. Duration of task-related fixations in the eyes, brow, nose and mouth regions was modulated by the presented expression. Moreover, longer fixation at the mouth positively correlated with anger and disgust accuracy both when these expressions were freely viewed (Experiment 2b) and when briefly presented at the mouth (Experiment 2a). Finally, an overall preference to fixate the mouth across all expressions correlated positively with anger and disgust accuracy. These findings suggest that foveal processing of informative features is functional/contributory to emotion recognition, but they are not automatically sought out when not foveated, and that facial emotion recognition performance is related to idiosyncratic gaze behaviour.

Characterization of facial emotion recognition in bipolar disorder: Focus on emotion mislabelling and neutral expressions.

Increasing evidence suggests that facial emotion recognition is impaired in bipolar disorder (BD). However, patient-control differences are small owing to ceiling effects on the tasks used to assess them. The extant literature is also limited by a relative absence of attention towards identifying patterns of emotion misattribution or understanding whether neutral faces are mislabelled in the same way as ones displaying emotion. We addressed these limitations by comparing facial emotion recognition performance in BD patients and healthy controls on a novel and challenging task. Thirty-four outpatients with BD I and 32 demographically matched healthy controls completed a facial emotion recognition task requiring the labelling of neutral and emotive faces displayed at low emotional intensities. Results indicated that BD patients were significantly less accurate at labelling faces than healthy controls, particularly if they displayed fear or neutral expressions. There were no between-group differences in response times or patterns of emotion mislabelling, with both groups confusing sad and neutral faces, although BD patients also mislabelled sad faces as angry. Task performance did not significantly correlate with mood symptom severity in the BD group. These findings suggest that facial emotion recognition impairments in BD extend to neutral face recognition. Emotion misattribution occurs in a similar, albeit exaggerated manner in patients with BD compared to healthy controls. Future behavioural and neuroimaging research should reconsider the use of neutral faces as baseline stimuli in their task designs.

Recognition of 3D emotional facial expression based on handcrafted and deep feature combination

• Deep and handcrafted features are extracted from 2D depth images and 3D scans respectively. • The extracted features are embedded into a covariance pooling for the dimensionality reduction. • Deep covariance learning is carried out using two additional layers to enhance their discrimination power. • The Gaussian kernel is used to generalize the radial basis function to the manifold space for SVM classification. • High classification performance is achieved on two challenging datasets. Facial emotion recognition (FER) methods have been proposed mainly using 2D images. These methods suffer from many problems caused by the difficult conditions of unconstrained environments such as light conditions and view variations. In this paper, we aim to recognize the emotional facial expressions independently of their identity using the 3D data and 2D depth images. Since the 3D FER is a very fine-grained recognition task, mapping the 3D images into 2D depth images may lack some geometric characteristics of the expressive face and decay the FER performance. Convolutional Neural Networks (CNN), however, have been successfully applied to the 2D depth images and improved handcrafted-based methods in computer vision and pattern recognition applications. For this reason, we combine in this paper two types of features; handcrafted and deep learning features and prove their complementarity for 3D FER. Favorably, covariance descriptors have proven a very good ability to combine features from different types into a compact representation. Therefore, we propose to use the covariance matrices of features (handcrafted and deep ones), instead of the features independently. Since covariance matrices belong to one of the manifold space types, formed by SPD (Symmetric Positive Definite) matrices, we mainly focus on the generalization of the RBF kernel to the manifold space for 3D FER using a supervised SVM classification. The achieved performance of the proposed method on the Bosphorus and BU-3DFE datasets outperforms similar state-of-the-arts.

The Extensive Usage of the Facial Image Threshing Machine for Facial Emotion Recognition Performance.

Facial emotion recognition (FER) systems play a significant role in identifying driver emotions. Accurate facial emotion recognition of drivers in autonomous vehicles reduces road rage. However, training even the advanced FER model without proper datasets causes poor performance in real-time testing. FER system performance is heavily affected by the quality of datasets than the quality of the algorithms. To improve FER system performance for autonomous vehicles, we propose a facial image threshing (FIT) machine that uses advanced features of pre-trained facial recognition and training from the Xception algorithm. The FIT machine involved removing irrelevant facial images, collecting facial images, correcting misplacing face data, and merging original datasets on a massive scale, in addition to the data-augmentation technique. The final FER results of the proposed method improved the validation accuracy by 16.95% over the conventional approach with the FER 2013 dataset. The confusion matrix evaluation based on the unseen private dataset shows a 5% improvement over the original approach with the FER 2013 dataset to confirm the real-time testing.

Facial emotion recognition performance influences executive control impairment in Anorexia Nervosa: an exploratory study

Objectives Since evidence on executive control among women with Anorexia or Bulimia Nervosa (AN/BN) are somehow inconclusive, we aimed to explore whether performance in set-shifting in AN/BN might be influenced by Facial Emotion Recognition (FER). Methods We randomly recruited women with a diagnosis of AN or BN, from an Eating Disorders Outpatient Clinic in Italy, as well as healthy controls (HCs). We evaluated with established tools: diagnosis (Eating Disorder Examination- EDE-17.0), executive control (Intra-Extra Dimensional Set Shift-IED) and FER (Ekman 60 Faces Test-EK-60F). Univariate distributions by diagnostic subgroups were assessed on sociodemographic and clinical variables, which were selected for subsequent multiple linear regression analyses. Results Women with AN performed significantly worse than HCs on IED adjusted total errors. HCs scored significantly better than AN and BN on EK-60F fear subscale. Although IED set shifting was associated (p = 0.008) with AN, after controlling for age, EK-60F fear subscale, alexithymia and depression (i.e., clinically relevant covariates identified a priori from the literature, or associated with AN/BN at univariate level), this association could not be confirmed. Conclusions Impaired executive control may not be a distinctive feature in women with AN, since several clinical characteristics, including fear recognition ability, are likely to have an important role. This has significant implications for relevant interventions in AN, which should aim at also improving socio-emotional processing.

Facial emotion recognition in borderline patients is unaffected by acute psychosocial stress

Borderline Personality Disorder (BPD) is characterized by difficulties in social cognition and social interactions, which exacerbate under stress. A previous study found better facial emotion recognition (FER) in patients with personality disorders and healthy controls (HC) after stress. We recently reported that emotional empathy scores, i.e. the emotional response to another person's emotional state, were significantly lower in BPD patients than in HC after psychosocial stress. Cognitive empathy scores remained unaltered. The present study aims to further investigate the effect of psychosocial stress induced by the Trier Social Stress Test (TSST) on FER as part of social cognition in patients with BPD. We randomized 43 women with BPD and 46 female HC to either the TSST or a placebo condition. Afterwards, participants were asked in an emotion recognition test to identify emotions in faces showing anger or sadness at low and high intensity. Both groups recognized emotions better at high intensity compared with low intensity. There was no effect of stress on FER performance and we found no difference between groups. This is in line with prior research on social cognition in BPD patients demonstrating that the ability to understand another person's perspective might be unaffected by acute stress.

Facial emotion recognition in adult with traumatic brain injury: A protocol for systematic review and meta-analysis.

Background:Traumatic brain injury (TBI) refers to head injuries that disrupt normal function of the brain. TBI commonly lead to a wide range of potential psychosocial functional deficits. Although psychosocial function after TBI is influenced by many factors, more and more evidence shows that social cognitive skills are critical contributors. Facial emotion recognition, one of the higher-level skills of social cognition, is the ability to perceive and recognize emotional states of others based on their facial expressions. Numerous studies have assessed facial emotion recognition performance in adult patients with TBI. However, there have been inconsistent findings. The aim of this study is to conduct a meta-analysis to characterize facial emotion recognition in adult patients with TBI.Methods:A systematic literature search will be performed for eligible studies published up to March 19, 2020 in three international databases (PubMed, Web of Science and Embase). The work such as article retrieval, screening, quality evaluation, data collection will be conducted by two independent researchers. Meta-analysis will be conducted using Stata 15.0 software.Results:This meta-analysis will provide a high-quality synthesis from existing evidence for facial emotion recognition in adult patients with TBI, and analyze the facial emotion recognition performance in different aspects (i.e., recognition of negative emotions or positive emotions or any specific basic emotion).Conclusions:This meta-analysis will provide evidence of facial emotion recognition performance in adult patients with TBI.INPLASY registration number:INPLASY202050109.

Semi-Supervised Learning for Continuous Emotion Recognition Based on Metric Learning

Emotion recognition is important for the interaction between human and artificial intelligence. Recently, the performance of facial image-based emotion recognition has been improved with deep learning's power. Nonetheless, huge data and label information for training are burdensome. In particular, annotating emotion labels in the continuous domain is very costly. Thus, we propose a novel semi-supervised learning that can not only reduce the annotation cost, but also improve emotion recognition performance by training with additional unlabeled data. The proposed method employs deep metric learning to improve feature embedding performance. Also, pseudo labels of unlabeled data are produced by analyzing inter-data distance in the feature space. Since pseudo labeling makes unlabeled data trainable, it increases overall performance. The experimental results show that the proposed method provides outstanding performance in the well-known MAHNOB-HCI dataset and the INHA dataset produced by our research team.

The Ability to Visually Recognize Facial Emotions during the Initial Stages of Alzheimer’s

Background: Alzheimer’s sufferers (AS) are unable to visually recognize facial emotions (VRFE). However, we do not know the kind of emotions involved, the timeline for the onset of this loss of ability to recognize facial emotional expressions during the natural course of this disease and the existence of any correlation with other comorbid cognitive disorders. For that reason, the authors aimed to determine whether a deficit in facial emotion recognition is present at the onset of Alzheimer disease, distinctly and concurrently with the onset of cognitive impairment or is it a prodromal syndrome of Alzheimer’s Disease before the onset of cognitive decline and what emotions are involved. A secondary aim was to investigate relationships between facial emotion recognition and cognitive performance on various parameters. Method: Single Blind Case-control study. Setting in Memory clinic. Participants: 12 patients, (AS) and 12 control subjects (CS) were enrolled. Measurements: Quantitative information about the ability for facial emotion recognition was obtained from Method of Analysis and Research on the Integration of Emotions (MARIE). The Mini Mental Status Examination (MMSE), the Picture Naming, the Mattis Dementia Rating Scale (DRS), and the Grober & Buschke Free and Cued Selective Reminding Test (FCSRT) tests were used to measure cognitive impairment. Results: We note that the AS have a problem with the visual recognition of facial emotions with existence of a higher threshold for visual recognition. The AS is less sensitive to the visual recognition cues of facial emotions. AS is unable to distinguish anger from fear. It would be a possible explanation for some acts of aggressiveness seen in the clinical and home setting demonstrated by “AS with behavioral disturbance”. The anger-fear series was found to be the first affected in the course of Alzheimer’s. The appearance of the curve is sigmoid for the control group and linear for the Alzheimer’s patients with a cognitive distortion when the VRFE is represented graphically with percentage of correct recognition plotted on the “y” axis and the selected images presented as stimulus with measures of density of emotion plotted on the “x” axis. In both groups, it is intuitively and theoretically expected that correct recognition will be directly proportional to the density of represented emotion in the stimulus image. This hypothesis is true for CS but not so for AS. The MARIE (see below) processing of emotions seems to be strengthened by the optimal cognitive functions showing the hypothesis applies to CS but not uniformly in AS. This anomaly in the AS is evidenced by the decline of the cognitive functions contributing to abovementioned “linearization” in the graphic representation. There is a direct positive correlation between the results of MARIE and the performance on cognitive tests. Conclusion: The administration of a combination of DRS, FCSRT, and MARIE to patients screened for possibly emerging Alzheimer’s could provide a more detailed and specific approach to make a definitive early diagnosis of Alzheimer’s. The Alzheimer’s patients found it difficult to distinguish between anger and fear.