Conscious and Unconscious Cognitive Processing: The Priming affect of Semantic Relatedness and Emotional Valance

by Michelle Kennedy for Cognitive Psychology

Abstract
This study explored conscious and unconscious processing in relation to word valence and mental

health (anxiety and happiness). A sample of 144 English speaking University students participated in the experiment. Batches of 20 students, grouped by valence, were administered three tasks (1) Semantic decision task on word pairs that included a positive, negative or neutral word plus a control; (2) Spielberg State-Trait Anxiety Inventory (SSTAI; Spielberger, Gorsuch, Lushene, Vagg,& Jacobs, 1983); (3)The Oxford Happiness Questionnaire (OHQ; Hills & Argyle, 2002). The Prime Relatedness main effect and Prime Relatedness by Word Type interaction were both significant, but the Word Type main effect was not significant. No significant effect sizes were found between mental health and reaction times. This study found both positive and negative valence do increase our unconscious processing speed especially when prime related, however neutral words resulted in little effect to priming. Further research is required to establish a link between mental health and cognitive performance.

Conscious and Unconscious Cognitive Processing: The Priming affect of Semantic Relatedness and Emotional Valance

Research into cognitive psychology has challenged traditional notions on perception, as studies conclude information processing can be unconscious (Merikle, Smilek & Eastwood, 2001). Yet as a rule, the extent to which stimuli is given attentive space depends on the importance of the message, which is considered a conscious process (Deutsch & Deutsch, 1963). Cognition relies on memory, perception, emotion, attention, associations and an individual lexicon (Deutsch & Deutsch, 1963; Eysenck & Keane, 2015; Okubo & Ogawa, 2013; Reisberg, 2013, Sass, Habel, Sachs, Huber, Gauggel & Kircher, 2012). Although some degree of cognitive arousal is essential for attention to operate, many believe processing occurs largely with conscious stimuli rather than unconscious stimuli and is a matter of selective attention (Deutsch & Deutsch, 1963). Despite this selective attention, no matter how alert we are there is a limit to the number of stimuli to which we consciously attend (Deutsch & Deutsch, 1963).

The question is to what extent do we process a subliminal stimulus, and how does emotion facilitate or hinder this process? Past research indicates that negative emotional arousal interferes with semantic processing and cognitive ability in subsequent tasks, which we will discuss (Sakaki, Gorlick & Mather, 2011). Studies using affective priming have successfully examined unconscious processing by masking semantically related primes (Okubo & Ogawa, 2013). Semantic priming is a well-known technique whereby a related word presented before a target word enhances processing of the subsequent target word (Heyman, Rensbergen, Storms, Hutchinson & DeDeyne, 2015), typically linked to a spreading activation process; masking thus renders the emotional valence of a word subliminal (Bodner & Masson, 2003). This priming effect has been proven to occur for word pairs that are associatively related, categorically related or simply look or sound similar (Okubo & Ogawa, 2013). The consensus of the research literature suggests that priming can be used in a variety of situations, including lexical decision-making and word labelling (Bodner & Masson, 2003). It will be argued that emotions play a vital role in cognition, with positive emotions enhancing cognitive abilities and that negative arousal is detrimental to cognitive processing as it reduces activation in the prefrontal cortex where perception, working memory, decision making and executive functioning is thought to reside (Itkes & Mashal, 2016).

Okubo and Ogawa (2013) conducted a study using affective priming to evaluate emotional valance (i.e., negative and positive) on target stimulus and observed RTs were faster when the prime and the target had the same valence. According to the Automatic Vigilance Hypothesis valence is automatic evaluated (Itkes & Marshal, 2016). A series of studies on unconscious processing, or perception without awareness found that information is perceived without attending or awareness of perceiving (Merikle, et al., 2001; Okubo & Ogawa, 2013), and adds further weight to an automatic processes. This defies common logic that we must perceive in order for information to affect thoughts, feelings and behaviour. Moreover, neuroscience has mapped the brain and found two pathways, which underlie emotional experience (Okubo & Ogawa, 2013), one of which may hold the secret to this unconscious processing. The cortical pathway (from the thalamus, through the cortex to the amygdala) is known as the conscious route and is evoked by attended stimuli. Whereas the subcortical pathway (direct thalamus-amygdala connection) may be evoked by subliminal stimuli and the emotional meaning appraised even if the stimuli has not been identified (Okubo & Ogawa, 2013). The studies conducted by Mack and Rock (1998; as cited in Merikle, et al., 2001), found unattended stimuli which participants claimed not to have observed, displayed measurable behavioural patterns associated with perception and thus concluded stimuli was perceived without awareness.

Deutsch and Deutsch (1963) assert that discrimination is required when attending, because as humans we find it difficult to attend to two stimuli simultaneously, such as talking on the phone and listening to a lecture. However, they found that unwanted messages were first processed before they were rejected or ignored, meaning that the two messages were first analysed for importance before one of them was discarded. Merikle et al., (2001) did a study to find out the difference, if any, in the consequences of conscious versus unconscious processing and found there were indeed qualitatively different consequences. The evidence supported perception without awareness under many conditions.

Implications for behaviour change from things, which arouse emotions, are huge; money for instance is highly emotive. Vohs, Mead, and Goode (2008) conducted a study on money and found merely activating the concept of money can change behaviour, for example participants subtly reminded of money preferred solitary tasks and were less helpful than those who were not reminded of money. Perception of a stimulus is personal and may be linked to ones mental health or valance of a stimuli; Yuan et al., (2011) completed studies on mood influence on brain functioning and found pleasant mood increases brain processing of cognitive control. In other words, it improved performance when the participants were required to inhibit task irrelevant activities from distracting them to achieve a goal. Further empirical evidence (Gray, 2001; as cited in Yuan et al., 2011) reveals mood acts as a modulating effect on the processes of cognitive control.

It is thought that due to the reduction in prefrontal cortex activation, negative valence captures attention automatically, more than positive words. Neuroimaging of the brain showed a reduced level of activation in the prefrontal cortex whilst conducting an automatic task. Sakaki et al., (2011) concluded that due to the considerable cognitive processing required for semantic connections, it is plausible that negative emotional states, such as anxiety which reduces activity in the prefrontal cortex, may interfere with semantic processing.

Against this background the aim of the present study was to examine whether the size of the priming effect in conscious and unconscious processing relates to word valence and if individual differences in happiness and anxiety play a role in the information we process. Therefore, it was hypothesised that prime related words with a positive valence will have faster reaction times (RTs) than control positive words. It was also hypothesised that prime related words with a negative valence will have slower RTs than control negative words. It was further hypothesised that prime related words with a neutral valence will have faster RTs than control neutral words. It was also hypothesised that participants scoring high on the OHQ and SSTAI will have faster RTs to negatively valence words than to positive or neutrally valence words.

Discussion

The purpose of the present study was to examine whether the size of the priming affect depends on word valence and if individual differences in happiness and anxiety affect information processing. We did find significant effect sizes for both positive words and negative words but not for neutral words. Analysis showed the direction of the reaction was negative for the emotionally valence words (positive and negative), indicating participants were indeed quicker on the primed words compared to the control words. The neutral word results were unexpectedly reversed; participants were quicker on the control pair than the prime pair. Although a weak positive correlation was found for individual differences of mental health the results were not significant as only 10% of the priming affect could be attributed to anxiety and only 8% to happiness. There is insufficient evidence to draw any conclusions.

The results of this study supported the hypotheses that prime related words with a positive valence did have faster RTs than control positive words. As discussed earlier there have been many studies, which show improvement in cognitive performance due to pleasant moods and positive valence (Sakaki et al., 2011) so this result is not surprising. An essential component of strong cognitive performance is the ability to organize information and inhibit task irrelevant thoughts. Spreading activation is notable, in that faster RTs are common with related concepts (Sass et al.,2012) perhaps positivity was used as a cognitive link. Research on attention is another possible explanation as results on positive moods indicate a broadening of attentive scope while a narrowing of attention occurs with negative moods (Sakaki et al., 2011). Therefore, our results for negative valence words are surprising.

Contrary to expectations, prime related words with a negative valence actually had faster RTs than control negative words. Furthermore, the negative primed pair resulted in even faster RTs than the neutral primed pair words. A likely culprit for this could be the link between the reduction in prefrontal cortex activity and automatic responding (Merikle et al., 2001). By shutting down the executive functioning, the individual was able to react in an instinctual manner rather than respond. Studies, which have monitored brain activity, found negative stimuli are given priority over neutral and positive stimuli (Itkes & Mashal, 2016). Deutsch and Deutsch’s theory of learning (1963, p. 85) predicts an inverse correlation where stimuli with high importance can be mistakenly perceived, therefore the increased RTs we obtained may simply be from link- analyser units in the brain making a mistake or allocating it as higher importance than other messages they were receiving at the time.

Our hypothesis, that prime related words with a neutral valence will have faster RTs than control neutral words was not supported. A possible explanation for these results could lie in the effects emotions have on our attention. Emotions are an essential component of human survival therefore it makes sense that emotional stimuli arouses attention faster and stronger than neutral stimuli, it has also been assessed as being more accurate and detected quicker than neutral stimuli (Sakaki et al., 2011). Spreading activation which assume that semantic activation spreads along the connections in long-term memory networks and makes associations which are fast and automatic (Bodner & Masson, 2003) fits this concept well.

Our final hypothesis, that participants scoring high on the OHQ and SSTAI will have faster RTs to negatively valence words than to positive or neutrally valence words was not supported. We didn’t find a significant connection to individual mental health over the relatedness of the prime.

As mentioned earlier the percentages than could be attributed to these emotional states were minimal.

Masked priming in the form of pre- and postmasking, prevented most subjects from perceiving their semantic properties, this was done to study unconscious affects of valence, however controversy around emotional arousal on cognition remains high. Our studies support the research thus cited even though a review by Mitchell and Phillips (2007; as cited in Yuan et al., 2011) on executive function and mood found opposing results, where positive mood inhibited performance during an Eriksen flanker task and negative mood had little effect on cognitive control. Rossell and Nobre (2004; as cited in Sass et al., 2012) who analysed affect on semantic priming also found no effect or inhibition for negative stimuli. They did however find a priming effect for positive and neutral stimuli. They postulated that neutral and positive stimuli might share some semantic properties therefore activating neural networks that weren’t available for negative stimuli.

This then brings up the limitations of this study; our research was confined to semantic stimuli and not procedural processing or cognitive control, which has found to have different outcomes for positive and negative stimuli compared with semantic results. Another limitation was that we used behavioural measures alone (e.g., response time) to discover the impacts of affective priming without neurophysiological measures. Therefore, we still don’t know what happens in the brain when confronted with emotional stimuli. Further research using both measures would be helpful to fully understand the complex processing that occurs when we perceive a stimuli whether it’s conscious or unconscious.

Despite these findings we lean towards the results our study found, that it is, emotions have a profound affect on our cognitive abilities and there is considerable difference in outcomes whether we consciously attend to something or perceive it subliminally. Positive emotions broaden our attentive scoop, tighten our semantic organization and generally enhance our cognitive performance. While negative emotions bypass our conscious and seem to be processed by our automatic response system. Negative arousal does disrupt some brain functions however it is usually noticed quickly and accurately over neutral and even positive stimuli. Our brain requires memory, attention, associations, and perception to process information but this can be an unconscious process just as easily as a conscious process. The implications of these findings for the general populous are substantial; first subliminal messages can create behaviour change and second negative stimuli is processed before neutral or positive stimuli and may decrease our executive functioning. The media is full of disturbing news and simply having it on in the background may seep into our brains and distort our behaviour. We need to be vigilant with what we allow into our lives and keep the tone upbeat for our own mental health and well-being if nothing else.

Some other theories which may be responsible for the priming effect are the spreading- activation theory, which assume that semantic activation spreads along the connections in long-term memory networks and makes associations which are fast and automatic (Bodner & Masson, 2003).

References

Bodner, G., & Masson, M. E. J. (2003). Beyond spreading activation: An influence of relatedness proportion on masked semantic priming. Psychonomic Bulletin & Review, 10(3), 645-652.

Deutsch, J. A., & Deutsch, D. (1963). Attention: Some theoretical considerations. Psychological Review, 70, 80-90.

Eysenck, M. W. & Keane, M. T. (2015). Cognitive psychology: A student's handbook (7th ed.). Hove, Sussex: Psychology Press.

Hills, P., & Argyle, M. (2002). The Oxford happiness questionnaire: A compact scale for the measurement of psychological well-being. Personality and Individual Differences, 33, 1073- 1082.

Itkes, O. & Mashal, N. (2016). Processing negative valence of words pairs that include a positive word. Cognition and Emotion, 30(6), 1180-1187, Doi: 10.1080/02699931.2015.1039934

Merikle, P. M., Smilek, D., & Eastwood, J. D. (2001). Perception without awareness: perspectives from cognitive psychology. Cognition, 79, 115-134.

Okubo, C. & Ogawa, T. (2013). Unconscious and conscious processing of negative emotions examined through effective priming. Psychological Reports: Relationships & Communications, 112(2), 607-625. Doi:10.2466/21.07.PR0.112.2.607-625 I

Residerg, D. (2103). Two types of priming. Cognition: Exploring the Science of the Mind, (5th Ed.). New York: Norton, 130-135.

Robinson, M. D.,& Kirkeby, B. S. (2005). Happiness as a belief system: Individual differences and priming in emotion judgments. Personality and Social Psychology Bulletin, 31(8), Doi: 10.1177/0146167204274081

Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R., & Jacobs, G. A. (1983). Manual for the state-trait anxiety inventory. Palo Alto, CA: Consulting Psychologists press.

Sakaki, M., Gorlick, M.A. & Mather, M. (2011). Differential interference effects of negative emotional states on subsequent semantic and perceptual processing. Emotion, 11(6), Dec 2011, 1263-1278. http://dx.doi.org/10.1037/a0026329

Sass, K., Habel, U., Sachs, O., Huber, W., Gauggel, S. & Kircher, T. (2012) The influence of emotional associations on the neural correlates of semantic priming. Human Brain Mapping, 33(3), 676-694. Doi: 10.1002/hbm.21241

Vohs, K.D., Mead, N.L., Goode, M.R. (2008). Merely activating the concept of money changes personal and interpersonal behavior. Current Directions in Psychological Science, 17(3), 208-212

Yuan,J., Xu,S., Yang, J., Liu, Q., Chen, A., Zhu,L., Chen, J., Li, H. (2011). Pleasant mood intensifies brain processing of cognitive control: ERP correlates. Biological Psychology, 87, 17–24. Doi:10.1016/j.biopsycho.2011.01.004

Leave a Reply

Your email address will not be published. Required fields are marked *