Attention and positive affect : temporal switching or spatial broadening?

Attention and Positive Affect: Temporal

Switching or Spatial Broadening?

A. Gül

Universiteit van Amsterdam

Tutors: Lotte Mulder & Hans Phaf Student number: 10595945

Word count: 4820

Abstract

Studies suggest a shared predisposition to preferentially direct attention towards emotional stimuli. Emotional effects on attention with non-emotional stimuli are less clear.The present study examined two opposing hypotheses focusing on this matter: the spatial- and the temporal hypothesis. The spatial hypothesis states that positive affect leads to a broader uptake of information, resulting in more interference of distractors and slower reaction time (RT). The temporal hypothesis states that positive affect enhances the flexibility of attention, resulting in less interference of distractors and quicker RTs.A visual search design was used to displays both mechanisms equally. Results showed a weak mood induction and little serial search. Analyses showed small effect sizes. No concrete conclusions were drawn. Further research is needed.

Attention and Positive Affect: Temporal Switching or Spatial Broadening?

Do our emotions affect the way we perceive the world? The interplay between emotion and cognition has been explored in a variety of subareas and the impact of emotions seems to be far reaching. For instance, it seems like people often tend to remember shocking circumstances extraordinarily vivid and in detail, such as the destruction of the World Trade Center in 2001 (Reisberg & Hertel, 2003). These memories are known as ‘flashbulb memories’. While it feels like these memories are very accurate, probably because of the strong emotions involved, they often turn out to be quite inaccurate (Reisberg & Hertel, 2003). Moreover, it has often been

observed that positive affect, which stands for the conscious subjective experience of feeling or emotion, boosts creativity (Zenasni & Lubart, 2002) and also influences reasoning, both with negative and positive emotional stimuli (Blanchette, 2006). As

stated earlier, many more cognitive functions are influenced by emotion. More than any other species we experience a richness of emotional events (Dolan, 2002), demonstrating the key role of emotion in our daily lives. Altogether, these emotional events indeed frequently affect the way we perceive the world (Zadra & Clore, 2011).

As humans we deal with opportunities and priorities on a daily basis (Phaf, 2015). We ascribe a value to the events happening during the day, which is a product of evolutionary selective processes (Dolan, 2002). This makes it possible for us to determine if an event or environment is more or less desirable (Dolan, 2002). The systems largely responsible for ascribing values to the events are often addressed as emotion and attention. Emotion and attention seem to be closely linked in a way that emotions can influence what pieces of information we grant priority for processing, such as in critical situations where it’s important to quickly shift your attention to the potential threat, like a loud noise or other arousing stimuli, so you can act upon it. As a matter of fact, emotions seem to facilitate both the speed (Öhman, Flykt, & Esteves, 2001) and the likelihood that arousing information is processed (Anderson & Phelps, 2001). When using emotional stimuli, such as pictures of people looking angry or sad, it has been found that there is more rapid target detection, i.e. more priority of

attention, for faces that express positive or negative expressions compared to neutral expressions. This effect also occurs with spiders and snakes, which are seen as evolutionarily relevant threatening stimuli (Öhman, Flykt, & Esteves, 2001) or as fear-relevant stimuli (Dolan, 2002).

A similar effect is found in Rapid Serial Visual Presentation (RSVP) tasks, which is known as ‘attentional blink’ (Raymond, Shapiro, & Arnell, 1992). In a RSVP task the purpose is to find a target in a continuous stream of visual items presented at the same place. When using non-emotional stimuli, the stimulus

presented after the target is often not seen. However, when the stimulus is presented after a target that consists of aversive, negative emotional content, it is less likely to be missed (Anderson & Phelps, 2001).

On the whole, the above-mentioned studies suggest that humans share a predisposition to preferentially direct attention towards emotional stimuli. However, it is less clear what the emotional effects on attention are when looking at

non-emotional stimuli.

A classic theory, known as the broaden-and-build theory (Fredrickson, 2004), suggests that positive emotions, such as joy and love, widen people’s thought-action repertoires. For instance, a thought-action repertoire could be joy stimulating the urge to play. It does so by enhancing an individual’s global scope, focusing more on the spatial effects, and increasing flexibility of attention, focusing more on the temporal effects (Fredrickson, 2004), integrating both the spatial and the temporal aspects of attention. This includes not only emotional stimuli, but non-emotional stimuli as well.

Instead of integrating the spatial and temporal aspects of attention, you can divide them into individual aspects, which leads to two contrasting hypotheses: a spatial- and a temporal hypothesis. By distinguishing both aspects, we can take a closer look at the emotional effects on attention with non-emotional stimuli.

Firstly, when focusing mainly on the spatial aspect of attention, or the spatial hypothesis, it has often been hypothesized that negative affect narrows the focus of attention, known as the ‘weapon focus effect’ (Loftus, Loftus & Messo, 1987). This effect implies having less attention for other details in the scene when, for example, there is a gun pointed at you (Kassin, Tubb, Hosch & Memon, 2001), which can lead to the details that catch your attention being narrowed down to peripheral details (Rowe, Hirsh & Anderson, 2007).

In a recent study by Rowe, Hirsh & Anderson (2007) the classic Eriksen flanker test was used to test for a proposed broadening function of positive affect. They found that positive affect leads to a decrease of visual selective attention by increasing processing of spatially adjacent flanking distractors which is being addressed to an increase of ‘scope’ of visual spatial attention (Rowe, Hirsh &

Anderson, 2007). This scope is also referred to as ‘attentional spotlight’. Attentional spotlight is the idea that attention moves around in our vision field so that the things falling in its ‘scope’ are being processed preferentially. Meaning, the study by Rowe, Hirsh & Anderson (2007) suggests that positive affects leads to a bigger scope of attention, which leads to more interference of the distractors, which they conclude from the fact that positive affect led to longer RTs in the Eriksen flanker task.

Earlier evidence for this scope assertion comes from studies that found that positive mood is associated with a more global focus of attention and that negative mood is associated with a more local focus of attention. In Gasper & Clore (2002) their experiments, it was found that positive mood promoted a greater focus on the forest and negative mood promoted a greater focus on the trees. Participants in a more negative mood were less likely to classify the given figures on global features (Gasper and Clore, 2002). In addition, participants in a more negative mood were also less likely to use the global concept as a guide for the reproduction of a drawing (Gasper and Clore, 2002).

However, when focusing more on the temporal aspect of attention, or the temporal hypothesis, contrasting results have been found. A study by Phaf (2015), which used a modified version of the Eriksen flanker task that was used in Rowe, Hirsh & Anderson (2007) their study, contrastingly found less interference of flankers when inducing positive affect, by adding a time element. In this study the appearance

of the flankers were limited in time by masks, which led to having less time to switch between flanker and target. The time between the flankers were varied which lead to longer RTs. This indicates more interference when smaller intervals were used with negative affect compared to positive affect. However, this effect turned around when longer time intervals were used, which suggests a temporal component of affective modulation of attention. All in all, it was found that there seemed to be less

interference of flankers with positive affect than with negative affect when shorter flanker-target intervals were used. This suggests that positive affect increases

flexibility (Phaf, 2015), because participants could switch faster between the flankers when in a positive affect.

The study by Phaf (2015) is not the only study that possibly tackles the spatial hypothesis. Another study that induced positive affect, which included three different experiments, failed to find evidence for a mood-induced widening of visual attention in all three experiments (Bruyneel et al., 2013). These contrasting results give us more reason to also look at potential studies supporting the temporal hypothesis.

For instance, a study by Dreisbach & Goschke (2004) showed that positive affect increased cognitive flexibility and reduced perseveration on a cognitive set-switching paradigm, which involves the ability to shift attention between one task and another (Dreisbach & Goschke, 2004).

Thus, the association made between positive affect and its impact on attention seems more complex than most theories hold. Given the current literature, there are two camps that contradict each other.

The tasks that were used so far in examining the relationship between

attention and emotion have put more emphasis on the spatial aspects, especially when using the Eriksen flanker task. This task requires individuals to view stimuli (typically

arrows) and selectively attend to a central target while ignoring irrelevant and distracting flankers (Eriksen & Eriksen, 1974). These distracting flankers are associated with an opposite response (“incongruent” = pointing in an opposite direction than the target), while facilitating flankers are associated with the same response as the target (“congruent” = pointing in the same direction as the target). These flankers are placed next to the target. Emphasis on the spatial aspect emerges, for instance, from the fact that the flanker interference decreases as the distracting flankers are further away from the center target (Eriksen & Eriksen, 1974). Because of having less emphasis on the time aspect, it is easier to find evidence for the spatial hypothesis, instead of finding support for the temporal hypothesis.

The goal of this study is to systematically distinguish both theories and examine whether positive affect leads to a broadening of visual attention or to an increase of flexibility of visual attention. Our solution to displaying both the

mechanisms equally is the use of a visual search task, which enabled us to include a spatial element: various sizes of concentric circles which the stimuli could be placed on, and to also add a time element: the RT when trying to find the target. To measure the time needed per stimuli presented, it is important to ensure serial search. Research has shown that when conjunctions of more than one separable feature are needed when distinguishing the possible stimuli, also known as the feature-integration theory of attention, attention must be directed serially to each stimulus when searching for the target (Treisman & Gelade, 1980). To establish serial search, distractors and targets in this experiment could vary in both orientation and form, i.e. conjunction search. This way we can pin both the temporal- and spatial hypothesis against each other to see what the effects of positive and negative affect are on our visual selective attention skills, when non-emotional stimuli are used.

Figure 1. The slopes we expect for both hypotheses when comparing positive- and negative

mood induction.

The experiment consisted of a visual search task, which was separated in two blocks: one after inducing positive affect and the other one after inducing negative affect. The order of mood induction was counterbalanced across participants. To induce these moods we let the participant choose form a list of sad songs and let them write a personal story about a negative event for inducing negative affect and, on a contrary, we let the participant choose from a list of joyful songs and let them write a personal story about a positive event for inducing positive affect.

We hypothesized that the reaction time will increase per added distractor. However, the slopes will be different for the negative mood induction and the positive mood induction (see Fig. 1). If the temporal hypothesis is true, the slope for positive affect will be less steep than the slope for negative affect, which means that in a positive mood participants will be quicker when detecting the target. If the spatial hypothesis is true, the opposite will be true, which means the slope for negative affect will be less steep than the slope for positive affect, which means that in a negative mood participants will be quicker when detecting the target.

Method

Participants

44 psychology students from the University of Amsterdam and friends and family of the experimenters participated in this experiment voluntarily, for course credit or 10 euros, after signing the informed consent. 10 of them were men and 34 of them were women. The ages ranged from 19 to 29 (M=21,93, SD=2,46). No specific exclusion criteria were applied.

Design

The visual search task had a 2 x 4 within-participants design. The first independent variable was the mood induction, which could be either positive or negative. The order of mood inductions were counterbalanced across participants by assigning the even number participants to first positive, then negative mood induction and by assigning the uneven number participants to first negative, then positive mood induction. There were four possible targets and each participant was assigned a target at random. The same target was used during the whole experiment. The second independent variable was the amount of stimuli presented. This amount could vary between four levels: 4, 10, 18 or 30 stimuli. Trial order was determined at random by the computer with a total trial amount of 480, practice trials not included. The

dependent variable in this study concerned reaction time (RT) to correct responses.

Materials

The stimuli were presented against a light-grey background on a monitor with a 1920 x 1080 resolution. Approximately, the distance from participant to the screen

was 60 cm. The music was played on MONACOR Stageline md5000dr headphones on an iPod Nano 16GB. The songs that were available are presented in Appendix A. The RTs were registered by a two-button response box. Participants were instructed to use their dominant hand to response when the target was present and their

non-dominant hand when the target was not present. Both the target and the distractors were presented in black and were 1-degree wide. The interval between two trials was 1000 ms with a random jitter of ± 500 ms to prevent rhythmic responses, represented by a black fixation cross. Participants were given 3000 ms/trial. The targets, if present, were positioned randomly on concentric circles of 3 different sizes of 5, 10 and 15 degrees wide and could vary in orientation and form (see example in

Appendix B). Moods were induced by letting the participant pick a sad or joyful song that they listened to while writing a positive or negative personal event and while doing the visual search task. Participants were given both a positive and a negative example story. At the start and after each mood induction and after each block, the mood of the participant was measured by the Self-Assessment Manikin (SAM), which is a non-verbal pictorial assessment technique (see Fig. 2). It was used to measure mood and arousal.

the bottom panel.

Procedure

Before the experiment started, participants were asked to read and sign the informed consent if they agreed with the terms. They were also given written instructions. In addition, the instructions were made clear verbally and participants were given the opportunity to ask questions if something was not clear. Participants were informed that the experiment investigated the influence of mood on reaction time when trying to detect a target between distractors and that this would involve both negative and positive mood induction and a simple visual search task. We emphasized that it was important for them to try to be as quick as possible and to really try to get in a positive or negative mood, depending on the mood induction. Before starting the actual experiment, participants were given the opportunity to get familiar with the task with twenty practice trials. If the participants reported to not fully understand the task yet or if the participant made more than 5 errors, they were given the opportunity to do twenty more practice trials. In the beginning of the actual experiment participants were asked to indicate their baseline mood level by indicating their pleasure and arousal in the SAM. After the SAM, the experiment started off with the first mood induction. Order of mood induction was determined on the basis of odd and even numbers of participants. Participants with an odd number were given

instructions for negative mood induction first and participants with an even number were given instructions for a positive mood induction first. For the mood induction, participants were instructed to write about either a positive or a negative personal event, while listening to a corresponding song of their choice from a list of five. The participants were explicitly told their story would not be recorded or used in any way

mood induction, a second SAM was conducted. After the SAM the participants could start with the actual trials. A total of 480 trials were divided in two blocks. After 240 trials the participants were asked to indicate their mood again with the SAM and were given a short break. Whenever they felt like resuming the task, they were asked to fill in the SAM, which was then followed by instructions for the second opposing mood induction. After picking a song and writing the story, participants were asked to fill in the SAM again, followed by the remaining 240 trials. Lastly, participants were asked to indicate their mood by filling in the SAM once more. At the end of the experiment an exit interview was held. In the exit interview participants were asked if their mood was too negative and if so, attempts were made to return it to a neutral mood state. They were also asked about the strategy they were using, if they felt any difference in reaction time and mood in the different mood inductions or if there was anything else that they wanted to share about the task.

Data Analysis Plan

To determine if the manipulation was successful, we will look at the pre-mood induction scores and compare these with the post-mood induction and post-block induction scores. We will do so by multiplying the pre-mood induction scores with two minus post-mood induction score and post-block score, for both the negative- and the positive mood induction. Subsequently, the positive mood induction score is subtracted from the negative mood induction score. A positive score will indicate a successful mood induction and a negative score will indicate a non-successful mood induction.

RTs will be compared by determining the mean RT scores of the participants after positive- and negative mood induction for 4, 10, 18 and 30 stimuli. The method

of least squares will be used to determine the slope and intercept of each mood induction separately to draw a conclusion about the hypotheses. Subsequently, we will look at the magnitude of the difference between the slopes through effect sizes.

The dominant statistical approach in psychology is Null Hypothesis Significance Testing (NHST), although it has many unappreciated problems (Nakagawa & Cuthill, 2007). For instance, NHST does not provide us information about the magnitude of an effect (Nakagawa & Cuthill, 2007), which we however are interested in. Also, it does not provide us information about the precision of the estimate of the magnitude of that effect (Nakagawa & Cuthill, 2007). NHST only gives information about the probability of the observed or more extreme data under the assumption that the null hypothesis is true, leading to two decisions: reject or fail to reject (Nakagawa & Cuthill, 2007). This approach encourages rejection or

acceptance rather than an approach that focuses on degrees of likelihood (Nakagawa & Cuthill, 2007), possibly missing out on valuable information.

Therefore, our emphasis will be on effect size when interpreting data in this study, which is one of the recommended practices besides NHST (Cumming, 2013).

Results

Differential moods were not successfully induced according to the

participants’ SAM reports. 27 participants showed no difference in mood, indicated by a negative score after the positive mood induction score was subtracted from the negative mood induction score of each participant. As a result, an eased criterion was applied for measuring differences in mood by subtracting post block scores after negative mood induction from the post block scores after the positive mood induction. Thus we could measure if participants were in a more negative mood after the

negative mood induction than they were after the positive mood induction. As with the previous criterion, a positive score indicated a successful mood induction and a negative score indicated a non-successful mood induction

After applying this eased criterion 13 participants reported no difference in mood, indicated by a negative score. These participants were excluded from further analyses. The average SAM scores of the remaining 31 participants are displayed in table 1.

Table 1

Average SAM scores of pre-mood induction, post-mood induction and post-block induction of the 31 participants that reported a successful mood induction.

Mood-induction

Pre-induction Post-induction Post-block

Positive 2,09 1,84 2,25

Negative 1,95 2.93 3,11

The order of the mood inductions did not reveal a reliable effect on any of the dependent variables in this study. The difference between the mean post-block SAM score after negative mood induction and the mean post-block SAM score after positive mood induction was 0,86.

After removal of the participants with a non-successful mood-induction 70,45% of the trials remained available for further analysis. The results of RTs on the visual search task are presented in Table 2.

Table 2

The RT results per mood induction of the participants with successful mood- induction. The results are displayed per array size of 4, 10, 18 and 30.

Induction Mean RT 4 Mean RT 10 Mean RT 18 Mean RT 30

Positive 611,06 (118,89) 673,35 (130,52) 726,75 (162,77) 744,5 (165,41)

Negative 651,64 (146,91) 712,24 (180,19) 753,73 (170,03) 772,63 (197,28)

These results show an increase in RT when array sizes become bigger, which is expected for both hypotheses. These RT results are displayed in figure 3.

Figure 3. Displaying the mean RT scores on array sizes of 4, 10, 18 and 30 stimuli of

participants with a successful mood induction.

550 600 650 700 750 800 4 10 18 30 R T i n ms Array size

RTs successful manipulation

These RTs were used to determine the best-fit straight line to the data, using the least squares method, providing us with the mean slope and the mean intercept displayed in table 3.

Table 3

The mean slope for both positive affect and negative affect with the mean intercept for the participants with a successful mood-induction.

Induction Mean slope Mean intercept

Positive 4,91 (4,28) 600,18 (121,69)

Negative 4,43 (4,39) 628,48 (137,54)

The results indicate a tendency towards the spatial hypothesis, as the slope for negative affect is less steep than the slope for positive affect. This can be derived from the fact that mean for the positive affect slopeis higher than the mean for the negative affect slope. However, the results show a very small effect size (95% CI [-1.72, 2.68], d = 0.11). Therefore, no firm conclusions can be drawn on the basis of these results. The slopes are displayed in figure 4.

Figure 4. Displaying the mean slopes of positive- and negative affect for participants

reporting successful mood induction.

Because several participants mentioned a pop-out effect in the exit interview and given the importance of serial search in this experiment, we removed their data and subsequently removed the data of participants that deviated 1 standard deviation from the total mean of RT’s to filter out the participants that seemed to have

experienced a pop-out effect as well. This led to a total of 10 participants. The RT results of the second analysis of the remaining 10 participants are displayed in Table 4.

Table 4

The RT results per mood induction of the participants with successful mood-

induction and more serial search tactics. The results are displayed per array size of 4, 10, 18 and 30. 600 620 640 660 680 700 720 740 760 780 4 10 18 30 R T i n ms

Slopes Successful Manipulation

Linear (Positive) Linear (Negative)

Induction Mean RT 4 Mean RT 10 Mean RT 18 Mean RT 30

Positive 637,96 (124,42) 721,94 (129,62) 837,60 (176,35) 873,28 (157,88)

Negative 682,76 (142,98) 761,57 (175,52) 837,10 (165,09) 894,44 (178,83)

These results show an increase in RT when array sizes become bigger, which is expected for both hypotheses. These RT results are displayed in figure 5.

Figure 5. Displaying the mean RT scores on array sizes of 4, 10, 18 and 30 stimuli of

participants with a successful mood induction and no-pop out effect.

These RTs were used to determine the best-fit straight line to the data, using the least squares method, providing us with the mean slope and the mean intercept displayed in table 5. 600 650 700 750 800 850 900 950 4 10 18 30 R T i n ms Array size

RTs no pop-out

Table 5

The mean slope for positive mood induction and negative mood induction with the mean intercept for the participants with a successful mood-induction and no pop out effect.

Induction Mean slope Mean intercept

Positive 9,10 (3,85) 626,63 (120,06)

Negative 7,97 (3,31) 670,39 (138,66)

The results indicate a tendency towards the spatial hypothesis too, as the slope for negative affect is less steep than the slope for positive affect. This can be derived from the fact that mean for the positive affect slopeis higher than the mean for the negative affect slope.

After carrying out the analysis without the pop-out participants, the results still showed a small effect size (95% CI [-2.24, 4.50], d = 0.31). Therefore, no firm

conclusions can be drawn on the basis of these results as well. The slopes are displayed in figure 6.

Figure 6. Displaying the mean slopes of positive- and negative affect for participants

reporting successful mood induction and no pop-out effect.

Discussion

The aim of this study was to experimentally distinguish between the spatial- and the temporal aspect of attention by using a design that displays both the

mechanisms equally. We hypothesized that the slopes of positive- and negative mood induction would differ. If the temporal hypothesis is true, the slope for positive affect would be less steep than the slope for negative affect. If the spatial hypothesis is true, the opposite would be the case. Because of the importance of successful mood induction and serial search, we conducted two analyzes representing the participants with successful mood induction and participants with possibly no pop-out effect as well.

As a result, an important concern is the size of the eventual sample sizes. In both analyses a significant number of participants were not included because of unsuccessful mood induction and no serial search, leading to a small sample size of

600 650 700 750 800 850 900 950 4 10 18 30 R T i n ms

Slopes No Pop-Out

respectively 31 and 10. Also, only small effect sizes were found for both analyses, with a tendency towards the spatial hypothesis: a negative effect slope that is a little less steep then the positive affect slope. Taking in account the small sample sizes and the corresponding small effect sizes, it is difficult to draw a conclusion regarding the two hypotheses.

Continuing on the failed mood inductions, a possible explanation could be related to the music in the experiment. A few participants reported to be quite distracted because of the music, causing them to pay less attention to staying in the appropriate mood. Also, another possible explanation is the general appreciation of music. This can lead to assessing music, even classified as ‘negative’, as pleasant and positive, meaning no substantial difference between both mood inductions.

As mentioned in the results, some participants reported a pop-out effect. This pop-out effect results in a non-serial way of searching. However, serial search is crucial in this experiment to determine how long it takes for a participant to find a target and if this differs in positive- and negative affect. A pop-out effect cuts away the time element in our experiment, which is important for determining the temporal representation of attention. When looking at average RTs in block 1 (M=729,04, SD=145,37 ms/trial) and block 2 (M=632,31, SD=107,06 ms/trial), in general, participants seemed to be quicker in the second block than in the first block, which indicates that reaction time improves across blocks. Although, it is just a small difference, which would indicate just a very small learning curve. Moreover, it could be a possibility that the participants became more familiar with the target during the experiment, which enabled them to see the target increasingly quicker and easier during the experiment, which then resulted in a pop-out effect. A solution to this problem would be to use varying targets during the experiment. This could result in a

more serial way of searching. Another point that requires consideration is the affect that the task could have been too easy. The average accuracy of all participants was very high (M=98%, SD=0,01), indicating that hardly any mistakes are made. A solution could be bigger array sizes with more stimuli presented at the same time. It could also be a possibility to add a separable feature, such as color. As stated before research has shown that when conjunctions of more than one separable feature are needed when distinguishing the possible stimuli, attention must be directed serially to each stimulus when searching for the target (Treisman & Gelade, 1980). However, it could be that the participants could too easily distinguish between the stimuli. Adding another separable feature could complicate the task, resulting in more serial search.

All in all, the mood induction was weak and the eventual sample sizes were small. The results indicated a tendency towards the spatial hypothesis, but the effect sizes were small. No concrete conclusions can be drawn from these results. However, the current literature does report findings for both hypotheses and an improved version of this study with better mood induction and a bigger sample size could give more clarification regarding which hypothesis is supported with this design.

References

Anderson, A. K., & Phelps, E. A. (2001). Lesions of the human amygdala impair enhanced perception of emotionally salient events. Nature, 411(6835), 305-309.

Blanchette, I. (2006). The effect of emotion on interpretation and logic in a conditional reasoning task. Memory & Cognition, 34, 1112–1125.

E. H. (2013). Happy but still focused: Failures to find evidence for a mood-induced widening of visual attention. Psychological Research, 77(3), 320-332. Cohen, J. (1992). A power primer. Psychological bulletin, 112, 155.

Cumming, G. (2013). The new statistics why and how. Psychological science, 0956797613504966.

Dolan, R. J. (2002). Emotion, cognition, and behavior. science, 298(5596), 1191- 1194.

Duncker (1945). On problem solving. Psychological Monographs, 58. Eriksen, B. A., & Eriksen, C. W. (1974). Effects of noise letters upon the

identification of a target letter in a nonsearch task. Attention, Perception, &

Psychophysics, 16(1), 143-149.

Fredrickson, B. L. (2004). The broaden-and-build theory of positive emotions.

Philosophical transactions-royal society of london series b biological sciences, 1367-1378.

Kassin, S. M., Tubb, V. A., Hosch, H. M., & Memon, A. (2001). On the" general acceptance" of eyewitness testimony research: A new survey of the experts.

American Psychologist, 56(5), 405.

Loftus, E. F., Loftus, G. R., & Messo, J. (1987). Some facts about" weapon focus.".

Law and Human Behavior, 11(1), 55.

Martin, E. A., & Kerns, J. G. (2011). The influence of positive mood on different aspects of cognitive control. Cognition and Emotion, 25(2), 265-279.

Nakagawa, S., & Cuthill, I. C. (2007). Effect size, confidence interval and statistical significance: a practical guide for biologists. Biological Reviews, 82(4), 591-605.

snake in the grass. Journal of experimental psychology: general, 130(3), 466. Phaf, R.H. (2015). Attention and positive affect: Temporal switching or spatial

broadening? Attention, Perception, & Psychophysics, 77, 713-719.

Phelps, E. A., Ling, S., & Carrasco, M. (2006). Emotion facilitates perception and potentiates the perceptual benefits of attention. Psychological science, 17(4), 292-299.

Gasper, K., & Clore, G. L. (2002). Attending to the big picture: Mood and global versus local processing of visual information. Psychological science, 13(1), 34-40.

Raymond, J. E., Shapiro, K. L., & Arnell, K. M. (1992). Temporary suppression of visual processing in an RSVP task: An attentional blink?. Journal of

experimental psychology: Human perception and performance, 18(3), 849. Reisberg, D., & Hertel, P. (Eds.). (2003). Memory and emotion. Oxford University

Press.

Rowe, G., Hirsh, J. B., & Anderson, A. K. (2007). Positive affect increases the breadth of attentional selection. Proceedings of the National Academy of

Sciences, 104(1), 383-388.

Treisman, A. M., & Gelade, G. (1980). A feature-integration theory of attention.

Cognitive psychology, 12(1), 97-136.

Zadra, J. R., & Clore, G. L. (2011). Emotion and perception: The role of affective information. Wiley interdisciplinary reviews: cognitive science, 2(6), 676-685. Zenasni, F., & Lubart, T. (2002). Effects of mood states on creativity. Current

Appendix A

Table containing the songs that the participant could choose from.

Table 1

The list of positive and negative songs that the participants could choose from.

Positive Negative

Pharrell Williams – Happy Eric Clapton – Tears in Heaven

Jackson 5 – ABC Adele – Someone Like You

Bob Marley – Don’t Worry Be Happy Sam Smith – Stay With Me Katrina & The Waves – Walking On Sunshine REM – Everybody Hurts The Jacksons – Blame It On the Boogie Passenger – Let Her Go

Appendix B

Trial example target absent and target present with 30 stimuli are represented in figure 7 and figure 8.

Figure 7. Trial example with target present.