Emotional Reasoning in the Human-Robot Interaction using Cognitive Robotics (PART-2)

By -

 

                                                        Image by montrealethics.ai
      

Methodology

Study design and participants

We employed a within-subjects experimental design to evaluate our hypothesis. Undergraduate volunteers enrolled in psychology classes at a nearby university took part in the study. The sample included 29 people (17 of whom were female; M=24.6 years, SD=2.42 years). The sample size was determined based on previous research that addressed emotional reactions comprehensively and had sample sizes ranging from 14 to 41. None of the subjects had a history of neurological or neuropsychiatric disorders, and none were using psychoactive medication at the time of the study. All participants signed a written informed consent form. The study was carried out following the most recent version of the Declaration of Helsinki (October 2013) and was authorized by the local university's Ethical Review Board.

Measurements

Baseline measures

Anxiety, motivation, and mood were measured at baseline to check for potential implications of dispositional and endogenous factors on participants' cognitive performances and emotional responses (Gendolla, 2000). The STAI was used to assess trait (20 items) and state (20 items) anxiety using the Italian version. The possible scores ranged from 20 to 80, with the greater the score, the higher the amount of anxiety. Internal consistency was satisfactory for both state (Cronbach's alpha=0.86) and trait (0.90) anxiety in this study. A 10-cm visual analog scale (VAS) was used to assess motivation and mood, ranging from 0 (not at all motivated to engage in this study/very bad mood) to 10 (extremely motivated/extremely good mood).

Valence

After each encounter, the Italian version of the Positive and Negative Affect Schedule was used to assess changes in positive and negative affect. It was created to offer quick measurements of positive and negative affect and comprises of two independent 10-item mood scales. On a 5-point scale (from 1='very slightly or not at all' to 5='very much'), respondents are asked to rate the amount to which they have experienced each emotion over a specific period. The PANAS can be used with a variety of timeframes, however, in this study, the timeframe utilized to measure state effect was 'at this moment.' Both the state positive (Cronbach's =0.83) and negative (Cronbach's =0.85) impact scales demonstrated strong reliability in the Italian version of the PANAS. With values of 0.76 and 0.89, Cronbach's alpha was satisfactory in this investigation.

Non-verbal behavior

All of the test sessions were videotaped for later study of the nonverbal behavioral patterns of each participant. An a priori coding system was established based on the Ethnological Interview Coding System (ECSI) to examine various elements of the subject's emotional and social functioning during two interviews. This study's coding scheme includes seven behavioral patterns that were divided into four major categories to represent people's emotional responses during clinical interviews. The coding method and definition of each behavior type were examined in this study. Two encoders were used to encode the video data. The duration or frequency of each behavioral pattern was determined based on its parameters. The percentage of time (length) spent staring at the doctor/robot over the entire testing session, for example, was measured as the "look" behavioral pattern. That is, the number of times (frequency) that the participant touched their face with their hands was counted to determine the hand-facial behavior pattern. We reduced the video rate to half that of real-time to improve the accuracy of the analysis. Before beginning analysis, encoders were trained to dependably encode behavioral patterns in the same way. Each encoder encoded three videos during training to verify appropriate stability before moving on to full data analysis. The intra-class correlation coefficient (ICC), which reflects the accuracy of the rating procedure, was calculated as a measure of interracial reliability.

Psychophysiological assessment


Bodyguard 2 (Firstbeat), which has been extensively utilized for HR measurement in both laboratory and ambulatory research, was used to continuously monitor heart rate using a conventional electrode setup. A visual inspection of successive R waves (identified by an automatic beat-detection method) was performed, and any irregularities were corrected.

HRV Analysis Software was used to derive time (root mean square of successive differences, RMSSD) and frequency (low-frequency HRV; LF-HRV, high-frequency HRV; HF-HRV, and LF/HF-HRV) domain HRV measures. The RMSSD reflects the integrity of vagus nerve-mediated autonomic control of the heart and is less susceptible to respiratory and movement artifacts, according to Task Force guidelines (Task Force of the European Society of Cardiology, 1996); the HF-HRV reflects parasympathetic activity; the LF-HRV is proportional to sympathetic activity but influenced by parasympathetic tone. The LF/HF-HRV ratio is typically used to calculate the LF/HF-HRV ratio, which is used to interpret LF-HRV principally as a measure of sympathetic tone.

 Cognitive workload


To guarantee that tasks in both contexts weighed similarly on participants' cognitive effort, the NASA Task Load Index was utilized to quantify the subjective workload experienced when engaging with the robot or the clinician. The NASA TLX assesses the overall workload incurred when using a particular program and determines the primary sources of workload, which are estimated across six dimensions: mental demand, physical demand, temporal demand, performance, effort, and frustration. Each dimension's perceived workload is graded on a 20-step bipolar scale. The weight of each workload task dimension is then assessed by comparing each dimension to the others. In each of the six scales, there are 15 possible pairwise comparisons. The number of times each component is selected is counted, and it might range from 0 (not relevant) to 5 (very significant) (more important than any other factor). The overall workload score is computed by multiplying each raw rating by the weight given to that element by the participant. The sum of the weighted ratings is then divided by 15 (the total weights) to provide an absolute workload score that runs from 0 to 100, with 100 representing the participant's highest effort.


Comments

Post a Comment

Please do not use foul language while commenting. We will appreciate your suggestions and reviews in the comments.

Read Also

Assisting Neuroimaging through DL

By -
Image
Introduction: In recent years, neuroimaging has been a research attraction, especially due to fMRI (Functional Magnetic Resonance Imaging) advancements. The fMRI data comprises of often several individuals. This data is used by researchers to study the association between the cognitive states of an individual and the underlying brain activities. The fMRI data used for neuroimaging is ideally suited for deep learning applications. The large, structured datasets of fMRI can be used for representation-learning methods of the DL. Generally, DL can be defined as a learning method with multiple levels of abstraction, where at each level the input data is transformed by a simple non-linear function which enables the model to recognize complex patterns. With higher-level representation, DL methods can associate a target variable with variable patterns in input data. Also, DL techniques can independently acquire these transformations from the data, eliminating the need for a comprehensive preex
Read more

Marine eDNA Analysis using DL techniques

By -
Image
Introduction Environmental DNA (eDNA) is the genetic information or material released by an organism into its surroundings through activities like skin shedding, mucus, urine or other secretions. These materials contain DNA fragments that can be found in soil, water or air samples. The concept of eDNA is based on the fact that organisms constantly leave organic traces behind them. By analysing the eDNA, researchers can detect the presence of any species in that area. This method is a valuable technique for biodiversity monitoring and ecological research.    Unlike traditional methods that involve direct observation or capturing of organisms, eDNA analysis allows researchers to study biodiversity without directly interacting with the species. This non-invasive approach is particularly useful for studying elusive or endangered species. eDNA analysis has been applied in various ecosystems, including freshwater, marine, and terrestrial environments. It is used to study aquatic organisms, t
Read more

Deep Neural Networks for ADMET properties' prediction

By -
Image
Introduction The absorption, distribution, metabolism, excretion and toxicity properties of any compound contributes to its potential as an orally administered drug. A set of criteria defined by Rule of Five or the Pfizer's Rule of Five stats that any compound meeting these criteria are more likely to have ADMET properties. The importance of these properties has increased in modern medicinal chemistry and new machine learning techniques are employed to predict these properties. The predictive ADMET models have helped in discovery of small molecules with improved safety and dose. Deep Neural Networks are showing usefulness in such predictions due to improved computational efficiency, larger datasets and adaption of image processing in the chemistry. Early Models The early DNNs employed showed improved prediction performance. Many researchers raised concerned that this improvement is in fact has no relation with better computational capacity, rather it is result of mere memorizing of
Read more

How Visual Cortex inspired the Convolutional Neural Networks

By -
Image
Introduction Since the 1960s, after the successful development (at least with respect to the existing technologies of that time) of the first perceptron by Frank Rosenblatt, researchers around the world have been trying to make computers more brain-like. In other words, they are constantly trying to develop programs and hardware that work much like our brains. However, creating a machine that thinks and programs like a human brain is still impossible, given the limitations of present technologies. But, we are successful in this direction up to an agreeable extent. The perceptron developed by Rosenblatt can be considered one of the first implementations of neural networks. It was a huge hardware model, rather than a Python program like today's perceptrons.  Rosenblatt's perceptron was a simple binary input-output system, with various limitations. It was not trainable for different patterns, rather but for simple image recognization. The major drawback was its inability to perfor
Read more