1 of 35

Stress

Raiyan Abdul Baten

Brooke Hamilton

2 of 35

Stress

“the non-specific response of the body to any demand for change”

Selye

3 of 35

4 of 35

5 of 35

Computed Stress

“the stress computationally derived from instantaneous measures of stress symptoms obtained by non-invasive methods.”

6 of 35

Physiological Measures

Skin conductivity
Heart activity
Brain activity
Blood pressure
Blood volume pulse
Muscle activity (EMG)
Skin temperature
Respiration

7 of 35

Physiological Measures

Skin conductivity

Galvanic skin response (GSR)

Flow of electricity through skin
Increased during stress

Empatica E4

Biopac GSR100C

Empatica

Mindfield

8 of 35

Physiological Measures

Heart Rate Variability (HRV)

Electrocardiogram (ECG or EKG)
Electrical activity produced by an impulse of ions flowing through cardiac muscles

9 of 35

Physiological Measures

Brain activity

Electroencephalography (EEG)
Electric waveforms at the scalp
Right hemisphere dominates during negative emotion

Emotiv EPOC+ headset

10 of 35

Physiological Measures

Blood pressure

Pressure exerted on the walls of blood vessels
Increases with increased stress

11 of 35

Physiological Measures

Blood volume pulse (BVP)

Photoplethysmography (PPG)

Measures the amount of light reflected by skin’s surface

Decrease in BVP correlated with increase in stress

iHealth

Cardiio app

12 of 35

Physiological Measures

Muscle activity

Electromyography (EMG)

Skin temperature
Respiration

13 of 35

Physical Measures

Behavior, gesture, and interaction

Body pose, body motion

Facial features

Facial expression
Eye gaze
Pupil dilation
Blink rates

Voice features

Increases in range and frequency
Fluctuations in frequency
Energy of high frequency components
Proportion of high frequency components

ASL-504 Eye gaze

Voice features can be used to measure the stress of drivers [reference 2 in second paper]. Increases in range and rapid fluctuations in fundamental frequency [101–103], increases in energy for high frequency voice components [104], and greater proportions of high frequency components

Murray et al. (1996) have reviewed various definitions of stress, and proposed a description of this phenomenon based on the character of the stressors. They have hypothesized four levels of stressors that can affect the *speech production* process. At the lowest level, these include direct changes on the vocal apparatus (zero-order stressors), unconscious physiological changes (first-order stressors), and conscious physiological changes (second-order stressors). At the highest level, changes can also be brought about by stimuli that are external to the speech production process, by the speaker's cognitive reinterpretation of the context in which the speech is being produced, as well as by the speaker!s underlying affective conditions (third-order stressors).

14 of 35

Measures of Stress

Invasiveness

Blood samples

Saliva samples

Urine samples

Pupil Diameter (PD)

Breathing pattern

Galvanic Skin Response (GSR)

Voice intonation

Body pose

Facial expression

Electroencephalography (EEG)

Blood volume pulse (BVP)

Heart rate variability (HRV)

Electromyography (EMG)

15 of 35

16 of 35

Computational Techniques

17 of 35

Artificial Neural Networks

S. Scherer, et al., Emotion recognition from speech: stress experiment, in: Proceedings of the 6th International Language Resources and Evaluation (LREC 2008), Marrakech, Morocco, 2008.

18 of 35

19 of 35

20 of 35

21 of 35

Artificial Neural Networks

Y. Fukuoka, A. Ishida, Chronic stress evaluation using neural networks, IEEE Engineering in Medicine and Biology Magazine 19 (2000) 34–38.

22 of 35

Lombard Effect

23 of 35

G. Zhou, et al., Nonlinear feature based classification of speech under stress, IEEE Transactions on Speech and Audio Processing 9 (2002) 201–216

24 of 35

Taegar Energy Operator

25 of 35

Hidden Markov Model

The SUSAS dataset: ”Speech Under Simulated and Actual Stress”

32 speakers uttered 16,000 isolated words in various stress conditions

HMM classifier to distinguish “stressed” vs “relaxed” based on TEO features outperformed everything else.

26 of 35

GREEN

27 of 35

RED

28 of 35

Yellow

29 of 35

Support Vector Machine

“Paced Stroop Test”

Inputs:

Galvanic Skin Response
Blood Volume Pulse
Pupil Diameter
Skin Temp

Output:

Classify “Stressed” vs “Relaxed” using SVM

30 of 35

Interventions

31 of 35

32 of 35

https://cleantechnica.com/files/2016/04/ModelS_Interior.jpg

In a 2 (driver emotion: happy or upset) x 2 (car voice emotion: energetic vs. subdued) experimental study, participants (N=40) had emotion induced through watching one of two sets of 5-minute video clips. Participants then spent 20 minutes in a driving simulator where a voice in the car spoke 36 questions (e.g., “How do you think that the car is performing?”) and comments (“My favorite part of this drive is the lighthouse.”) in either an energetic or subdued voice. Participants were invited to interact with the car voice. When user emotion matched car voice emotion (happy/energetic and upset/subdued), drivers had fewer accidents, attended more to the road (actual and perceived), and spoke more to the car. Implications for car design and voice user interface design are discussed.

1 of 35

2 of 35

3 of 35

4 of 35

5 of 35

6 of 35

7 of 35

8 of 35

9 of 35

10 of 35

11 of 35

12 of 35

13 of 35

14 of 35

15 of 35

16 of 35

17 of 35

18 of 35

19 of 35

20 of 35

21 of 35

22 of 35

23 of 35

24 of 35

25 of 35

26 of 35

27 of 35

28 of 35

29 of 35

30 of 35

31 of 35

32 of 35

33 of 35

34 of 35

35 of 35