Hello Jim,

Well I'm not an expert, but I think SCR method is not good for such thing. SCR is depended from many things such as skin condition during measurement, which isn't stable and is changing during time (especially in the places where electrodes are attached). You can find many articles about EOG where SCR is one of main obstacles. I think EEG will be better suit for your research.

But, maybe I'm wrong and SCR will give some interesting results. The best way will be combine EEG with SCR and compare results, which you can publish later.

Best regards,

Jacek Chodak

Hi Jim, we performed a number of experiments while using combinations of SCR, EEG, EOG to measure muscle artefacts,  HRV, breathing frequency and in some cases an eye tracker. It is possible to perform such measurements but you will get a challenge analysing and explaining the results. Check some publications of my colleague Dr. A.M. Brouwer on this matter she has experiences.

Hi,

SCR/SCL reflects sympathetic nervous system activity. Situations that cause increased sympathetic activity usually catch people's attentions and typically are "emotionally salient". Increased SCR/SCL can be observed in so many conditions (e.g., viewing positive and negative pictures; making errors; surprise; losing/winning money) thus people tend to think increases in SCR/SCL does not link to any specific types of stimuli. This is mostly true. However, it will make more sense if you think about that increased sympathetic activity can happen in many different situations (e.g., now I am typing to reply this message). You may check Don Fowles's research for more mechanistic meanings of SCR/SCL. 

To address individual differences in skin conditions and overall electrodermal activity, some methods have been developed. A commonly used one is as followed:

Lykken, D. T., & Venables, P. H. (1971). Direct measurement of skin conductance: A proposal for standardization. Psychophysiology, 8(5), 656-672. doi:10.1111/j.1469-8986.1971.tb00501.x

Hope those help. 

-Kuan

HI,

The response is mainly modulate4d by an increase in sympathetic activity and is also about two seconds delayed.  Depending on the person, you can attend to stimuli without evoking a SCL response.  In our educational work, a slightly more sensitive system is blood volume pulse amplitude which will usually decrease if the attention activates some component of fear or self-judgement. 

If the stimulus is repeated and repeated visually and it grabs your attention then EEG alpha blocking can be a marker of how you were captured by the stimulus.  See my article,

Peper, E. (1970). Feedback Regulation of the Alpha Electroencephalogram   Activity Through Control of the Internal and External   Parameters.  Kybernetik, 7(3), 107-112.

Skin conductance is a marker of emotional impact (as mentioned by colleagues above). As emotions can be unconscious, and unconscious information is not attended, then skin conductance is not correlated with attention.

Another problem with SCR besides its non-specifity includes its time delay. The repsonse unfolds over a few seconds, peaks and than vanishes again until baseline is reached it takes about 7 seconds. So what you usually get is many overlaid responses. You might circumvent this problem by using ledalab or a similiar algorithm to directly indicate nerve activity from the SCR stream this gives you a sufficient temporal resolution comparable to eeg measures. I combined it with eyetracking in my diploma thesis and it works fine - however not without limitations. Attached you find a copy where we did it with EMG to indicate emotion.

Conference Paper Do You Like what You See? Combining Facial EMG and Eyetracki...

Jim, Following article compares the most specific way one can distinguish emotions. Respiratory pattern in combination with electrodermal responses seems to be the best.

 J Auton Nerv Syst. 1997 Jan 12;62(1-2):45-57.

Autonomic nervous system response patterns specificity to basic emotions.

Collet C1, Vernet-Maury E, Delhomme G, Dittmar A.

Author information

Abstract

The aim of this study was to test the assumption that the autonomic nervous system responses to emotional stimuli are specific. A series of six slides was randomly presented to the subjects while six autonomic nervous system (ANS) parameters were recorded: skin conductance, skin potential, skin resistance, skin blood flow, skin temperature and instantaneous respiratory frequency. Each slide induced a basic emotion: happiness, surprise, anger, fear, sadness and disgust. Results have been first considered with reference to electrodermal responses (EDR) and secondly through thermo-vascular and respiratory variations. Classical as well as original indices were used to quantify autonomic responses. The six basic emotions were distinguished by Friedman variance analysis. Thus, ANS values corresponding to each emotion were compared two-by-two. EDR distinguished 13 emotion-pairs out of 15. 10 emotion-pairs were separated by skin resistance as well as skin conductance ohmic perturbation duration indices whereas conductance amplitude was only capable of distinguishing 7 emotion-pairs. Skin potential responses distinguished surprise and fear from sadness, and fear from disgust, according to their elementary pattern analysis in form and sign. Two-by-two comparisons of skin temperature, skin blood flow (estimated by the new non-oscillary duration index) and instantaneous respiratory frequency, enabled the distinction of 14 emotion-pairs out of 15. 9 emotion-pairs were distinguished by the non-oscillatory duration index values. Skin temperature was demonstrated to be different i.e. positive versus negative in response to anger and fear. The instantaneous respiratory frequency perturbation duration index was the only one capable of separating sadness from disgust. From the six ANS parameters study, different autonomic patterns were identified, each characterizing one of the six basic emotion used as inducing signals. No index alone, nor group of parameters (EDR and thermovascular for instance) were capable of distinguishing each emotion from another. However, electrodermal, thermo-vascular and respiratory responses taken as a whole, redundantly separated each emotion thus demonstrating the specificity of autonomic patterns.

 Thanks everyone for the very useful and helpful responses. It sounds like there may be other physiological markers of attention and emotion that could be useful to investigate. However, we do feel SCR has some merits for what we want to do, i.e. act as a marker of visual attention. I have found this previous study which uses SCR in similar way to the approach we have been thinking about, to indicate perception of hazardous situations in a driving simulator:

http://www.sciencedirect.com/science/article/pii/S1369847814000540

Thanks again for your advice.

Jim

http://www.sciencedirect.com/science/article/pii/S1369847814000540

Hi Jim,

depending on the nature of your stimuli you might be able to get a response in SCR. Please note however that in our simulator study the hazardous situations appeared "dangerous" to the subjects and we therefore got a strong emotional response.

You have to be aware of the timing limitations as well. The response in SCR won't be instantaneous (as the eye tracking signal is).

 Thanks for your response Thomas. Our proposed study would be exploratory in nature so yes, one issue may be whether there are enough things going on that are hazardous or emotionally salient enough to evoke a SCR. I guess we won't know until we try! Your study gave us some hope though.

No !

SCR is a measure of emotional arousal.

You need to use eyetracking with behavioural response.