EDR Interpretation and Analysis in Naïve and Primed Subjects

Abstract

Electrodermal Response (EDR) remains a valuable resource for measuring the physical manifestations of emotions. To measure the effects of existing emotional state and prior knowledge on the results of an EDR test, this study asked nineteen Experimental Psychology students a series of twenty emotional and neutral questions. The participants were broken into two groups: primed and naïve. The study compared the height of each participant’s EDR from the primed group against the peak EDR from the participants in the naïve group. The study sought to show that a participant from the naïve group would have a stronger EDR to the emotional questions than a participant from the primed group, and secondarily that emotional questions would create a stronger EDR than the neutral questions. The results show that the primed group had a significantly stronger response to the questions than the naïve group and that there was a negligible difference in EDR between the emotional and neutral questions.

EDR Interpretation and Analysis in Naïve and Primed Subjects

As psychology and neuroscience still push to discover the connection between psychological and physical reactions in the body, and how conscious these reactions are (Critchley & Nagai, 2012), studying the effects of conscious awareness of Electrodermal Response (EDR) can be a useful tool. Because of the nature of the palmar readings necessary for EDR, this type of test delivers a more accurate reading of a subject’s emotional state, making it a reliable source for measuring the bodily effects of changed emotions. EDR can be an important tool for measuring how anxiety disorders affect the body during episodes of anxiety or panic and discovering what unconscious reactions occur within the subject (Thayer, Friedman, Borkovec & 1996).

Since facial expressions and physical behavior can be unconsciously changed by outside sources (Kret, Stekelenburg, Roelofs & de Gelder, 2013), EDR can provide a solution to outside influence by directly reading the body’s unchangeable chemical reactions to emotion. It takes away the room for misinterpretation inherent in relying on visual cues the subject makes, consciously or otherwise, and instead records the direct reactions from the body in a straightforward manner. This may prove particularly useful in studying patients with affective disorders.

The ability to gauge what the connection between prior awareness and the effects on the body could lead to a better understanding of how to find triggers for affective disorders, or how those disorders influence the functions of the body. It would mean being able to distinguish between a reaction due to the affective disorder and one that comes solely from surprise at hearing a question designed to evoke a reaction for the first time during a testing session. If this experiment were able to show that removing the element of surprise and unfamiliarity from questioning for an EDR, then researchers could prime their subjects and receive more accurate results.

This experiment was to test the relationship between primed and naïve subjects during an EDR reading. It hypothesized that a naïve subject asked a series of questions would have a higher reaction than a primed subject, and secondly, that emotional questions would generate a higher EDR response than neutral questions.

Method

Participants

The participants were 19 Experimental Psychology students, 4 (21%) males and 15 (79%) females. These students were chosen as a convenience sample. The students were of college age, between 18-21. Students within the Primed sample were acquainted with the equipment and the experimental process prior to performing the experiment on themselves, as guided by a class instructor. The Naïve students were admitted to the testing area after the demonstrations were over and thus were not acquainted with the experiment or its goals.

Materials

The EDR was conducted with the Power Lab Teaching System apparatus from ADInstruments (PTB4246); Data was collected from the apparatus with a multi-channel bio amplifier and EDR finger electrodes and entered into the Lab Chart software included with the apparatus. The software ran on an iMac 64-bit desktop computer. The 20 questions (10 emotional, 10 neutral) posed to the subjects during the EDR readings were adapted on the Lab Chart software from a different prior experiment.

Procedure

Participants were sorted into two groups: Primed, and Naïve. The Primed group consisted of students present for a demonstration of the Power Lab Teaching System apparatus and thus were privy to the proposed hypothesis. The Naïve group consisted of students outside of the demonstration group, and thus who had no prior knowledge of the apparatus, or of the hypothesis.

The subject would sit faced away from the computer screen, and have the finger cuffs attached to their clean hands. The subjects were instructed to remain relaxed, and answer honestly, as their responses would be confidential. They sat with their palms facing upwards. The researchers did not begin until the subject’s EDR became steady at +/- 0.10μS. The subject was asked questions at about ten-second intervals to ensure the EDR returned to its baseline; the emotional questions and neutral questions were alternated, with the peak EDR response being recorded.

The neutral questions asked to the subject included the following: Do you live in a dorm; Do you like broccoli; Do you have a cat; Are you a college senior; Have you ever ice skated; Do you ride a bike to school; Have you ever been to Alaska; Do you have a sister; Were you born in the US; Are you a psychology major.

The emotional questions included the following: Have you ever been in love; Do you ever cry; Does nudity embarrass you; Do you remember your first kiss; Does being alone in the dark frighten you; Have you ever seen a fatal accident; Do you know people that use drugs; Do you ever lie; Should safe sex be taught to middle school students; Have you ever seen a dead body.

Each set of questions was designed to achieve a certain effect. The emotional questions strove to evoke an emotional response by pursuing topics related to emotions or strongly held beliefs, while the neutral questions were merely yes or no questions with no emotional connotation.

Results

The results show that the primed group of subjects had a significantly higher reading than the naïve subjects. The naïve neutral subjects M = 1.55 (SD = 1.18), and the naïve emotional subjects M = 1.58 (SD = 1.22). Within the primed group, the neutral M = 4.54 (SD = 3.50), and the emotional M = 4.67 (SD = 3.70) (see Tables 1 & 2). The naïve test showed a correlation of 0.99 between the Neutral and Emotional responses, and a correlation of 0.99 in the Primed test.

The one-way analysis of variance (ANOVA) test was performed as a 2x2 factorial design. The ANOVA results showed that there was significance between the Primed and Naïve emotional groups, r(16) = 5.84, p = <.029 (see Figure 1). The ANOVA test also showed significance between the Primed and Naïve neutral groups, r(16) = 5.59, p = <.032 (see Figure 1). This confirms that there is a significance between the Primed and Naïve groups. The graph shows that there is no significance between emotional and neutral questions.

Discussion

The results show that both portions of the hypothesis were incorrect. The first hypothesis, which stated that the Naïve group would have a higher emotional response, was contradicted by the findings that showed the Primed subjects had a stronger reading. The second showed that the neutral and emotional questions had no significance, while the hypothesis proposed that there would be a stronger reading for the emotional questions. Thus, we can conclude that priming has a significant effect on EDR, more even than emotionally stimulating questions.

This result is a little surprising, considering the nature of the priming; the students were not consciously put into an emotional state prior to the beginning of the testing, which suggests there may be a connection between the subjects’ expectations and EDR. Murphy and Zajonc (1993) conducted an experiment to find a connection between priming and unconscious influence; the study found that showing subjects blurred Chinese ideographs made subjects unconsciously drawn to clearly visible versions of the blurred ideographs over completely new ideographs. The results of the 1993 experiment suggest that when the subjects of our experiment were affected by their knowledge of the hypothesis, even without knowing that they were affected, leading them to experience heightened emotion as they were tested.

The general unresponsiveness to the emotional versus neutral questions is also surprising. This reaction may be explained by the study conducted by Critchley and Nagai (2012) on bodily states affecting emotions. They say that “bodily states shape mental content,” and describe their experiment, in which they inject subjects with either saline or adrenaline and put them into a room with a “stooge” to evoke an emotional response from the subject (Critchley & Nagai, 2012, p. 163-4). The study showed that patients injected with adrenaline had stronger emotional responses than the subjects injected with saline (164). Given the results of Critchley and Nagai’s study, it is possible to infer that the primed subjects of our study experienced their own version of the adrenaline injection, and their bodies and consciousness reacted in a way the naïve subjects did not.

Critchley and Nagai also concluded that when participants of a study had to hold a pen with their lips and react to cartoons, the inability to smile naturally (restricted by the pen) lead them to perceive cartoons as being less funny than subjects who held a pen with their teeth, and thus were able to react naturally (2012, 164). Though there was no physical restriction on the subjects of our study, the connection between physical influence to experienced emotional stimulus may have some relevance to our findings; if our subjects could not feel emotions “naturally,” or in a context which allowed for the encouragement of emotional responses, the lack of higher emotional EDR from both the Primed and Naïve groups may be explained.

The Mood-Behavior Model (MBM) suggests that perhaps because there was no reason to react behaviorally to the emotional questions, the subjects may not have had an appropriate stimulus to evoke strong emotional reactions (Gendolla, 2000). Gendolla’s study (2000) also emphasizes the importance of moods as priming factors, meaning that the subjects may not have been in a congruent mood to evoke the necessary emotions to generate a high EDR for our study. Kret et al. (2013) emphasize the importance of interpreting outside factors in developing emotional stimulus, which perhaps confirms the idea that there was not enough appropriate outside stimulus for the subjects to develop the high emotional EDR expected by our hypothesis.

Unfortunately, this particular experiment does not seem to offer any helpful results regarding its applications for studying affective disorders. Thayer et al. (1996) proposed that there is a connection between general anxiety disorder (GAD) and inhibition in EDR responses compared to non-anxious subjects; since our study found only a possible connection between knowledge as a primer and high EDR responses, it seems unlikely that the results can offer anything to further or dispute any information about how or why GAD inhibits EDR responses, or how to better use priming to further study on GAD with EDR.

Similarly, subjects with unipolar and bipolar depressive disorders had universally inhibited EDR results (Iacono, Lykken, Peloquin, Lumry, Valentine & Tuason, 1983). Iacono et al. discovered that the inhibited EDR results were a reliable indicator for depressive disorders, but our study had inhibited responses in all categories except for priming. There is no obvious connection between our results and the results from Iacono et al.’s (1983) study.

Because our EDR results did not indicate positive or negative emotion, it is difficult to apply the results to discovering the direct causation of those emotions. Though it focused on stressors contributing to early cardiovascular disease, Gallo, Smith, and Kircher’s (2000) study found that social contexts and interpersonal relationships contribute to how a subject reacts to stressors; this may suggest that because the Primed group—the group with the highest EDR readings—had previous relationships in their classroom setting, they may have been more prone to stressors during the questioning process. This would have created a higher peak in EDR readings than for the Naïve group, who were unfamiliar with the hypothesis, and were comparative outsiders to the Primed group performing the questioning.

In a study on the effect of the media used to stimulate EDR subjects during experiments, Simons, Detenber, Roedema, and Reiss (1999) discovered that the medium used to deliver the stimulus had an effect on the subject’s reaction. Simons et al. focused primarily on images and motion in their study, but it presents the possibility that the medium through which the subjects were questioned may have been inappropriate for this particular experiment. There may have been a disconnect between the participant and the questions, meaning the readings were inaccurate.

One possible flaw in this experiment lies with the emotional questions in particular; if the questions simply failed to evoke an emotional response because the subjects had no personal emotional connection to the topics chosen, that would affect the results and the conclusions of this study. Gunes and Pantic (2010) show that there are many more layers to emotion that were not addressed by this study (valence dimension, arousal dimension, and power dimension), which may have resulted in a deeper understanding of the emotional connection or disconnect from the questions, and why the emotional readings were so low.

Since the study did not specify a participant pool of students with affective disorders, it had no direct way of making a connection between the results and the possibility of furthering the study of affective disorders and EDR. Had the initial hypothesis been correct, perhaps it would have led to further study with a participant pool with affective disorders to discover whether priming the participants to the study may have led to more accurate readings, or whether the shock of unfamiliar stimuli was affecting the EDR responses of those participants in a way that changed the overall outcome of the study.

The study may benefit from a larger pool of participants, particularly with a larger gender distribution. If it seeks a practical application, it might also find participants that are diagnosed with or suspected to have affective disorders, to see how knowledge priming affects the inhibited results associated with anxiety and depressive disorders. A wider variety of questions may be helpful in assuring a stronger emotional response from the participants, especially if the questions address a broader range of emotions; adding in more questions that have stronger ties to controversial social and political topics may eliminate the possibility that the questions were flawed and failed to elicit the reactions they were developed to evoke, rather than the participants feeling no emotional stimulus in response.

As it stands, this study has some interesting conclusions about the effect of knowledge priming on EDR readings. Since there seems to be a lack of similar studies, it would be beneficial to push forward in an attempt to study knowledge-priming further, especially with different participant pools. Since the experiment was conducted with limited questions and participants, it does not currently connect in a meaningful way with other existing studies on EDR and affective disorders or emotional readings. The study has not yet reached its full potential, and it must be opened to wider audiences before it begins to truly emerge.

(Figure 1 omitted for preview. Available via download) 

References

Critchley, H. D., & Nagai, Y. (2012). How emotions are shaped by bodily states. Emotion Review, 4(2), 163-168. Retrieved November 11, 2013, from http://emr.sagepub.com/content/4/2/163

Gallo, L. C., Smith, T. W., & Kircher, J. C. (2000). Cardiovascular and electrodermal responses to support and provocation: Interpersonal methods in the study of psychophysiological reactivity. Psychophysiology, 37(3), 289-301.

Gendolla, G. H. (2000). On the impact of mood on behavior: An integrative theory and a review. Review of General Psychology, 4(4), 378-408.

Gunes, H., & Pantic, M. (2010). Automatic, dimensional and continuous emotion recognition.

International Journal of Synthetic Emotions, 1(1), 68-99.

Iacono WG, Lykken DT, Peloquin L, Lumry AE, Valentine RH, Tuason VB. Electrodermal activity in euthymic unipolar and bipolar affective disorders: A possible marker for depression. Arch Gen Psychiatry.1983;40(5):557-565. doi:10.1001/archpsyc.1983.01790050083010.

Kret, M. E., Stekelenburg, J. J., Roelofs, K., & Gelder, B. d. (2013). Perception of face and body expressions using electromyography, pupillometry and gaze measures. Frontiers in Psychology, 4. Retrieved November 14, 2013, from http://www.frontiersin.org/Cognitive_Science/10.3389/fpsyg.2013.00028/abstract

Murphy, S. T., & Zajonc, R. B. (1993). Affect, cognition, and awareness: Affective priming with optimal and suboptimal stimulus exposures. Journal of Personality and Social Psychology, 64(5), 723-729.

Simons, R. F., Detenber, B. H., Roedema, T. M., & Reiss, J. E. (1999). Emotion processing in three systems: The medium and the message. Psychophysiology, 36(5), 619-627.

Thayer, J. F., Friedman, B. H., & Borkovec, T. D. (1996). Autonomic characteristics of generalized anxiety disorder and worry. Biological Psychiatry, 39(4), 255-266.