However, similar results were not seen for the time to first hit attention orientation in our eye tracking data. One possible explanation for this is that because of the numerous visual inputs participants received, they tried to obtain an overview in the beginning by gazing at all faces but then reacted faster to gain targets following gain cues even if they had only paid peripheral attention to those targets. However, more focused attention may subsequently have been diverted to gain targets following gain cues during later stages of attention. Consistent with such a view, the results for the percentage of gazing at the target half a second after the first hit attention maintenance were very similar to the effects seen in RTs.
Participants looked more at gain targets when they had optimistic expectancies compared with pessimistic expectancies and optimistic expectancies made participants look more at gain targets compared with loss targets within a half second after the first hit at a target. Similar to the RTs, the percentage of looking at gain and loss targets half a second after the first hit did not differ when pessimistic or ambiguous expectancies were induced. Moreover, the percentage of looking at loss targets did not differ when optimistic expectancies were induced in comparison with pessimistic expectancies.
In conclusion our cue and target congruency hypotheses could only be confirmed for optimistic expectancies, not for pessimistic expectancies. Notably, although in our first hypothesis we had predicted that pessimistic expectancies guide attention toward punishment compared with reward, this result is congruent with our second hypothesis that optimistic expectancies have a stronger influence on subsequent attention to reward and punishment than pessimistic expectancies do. A similar effect was seen in our eye tracking measure for attention maintenance.
Optimistic expectancies made participants look more at gain targets compared with loss targets than pessimistic expectancies made participants look more at loss targets compared with gain targets half a second after the first hit. However, the trend for both effects was non-significant. Therefore, whether optimistic expectancies had a stronger effect on attention deployment to congruent confirming compared with disconfirming information than pessimistic expectancies was not clearly shown in our data and requires further investigation.
In summary, our cue and target congruency hypotheses were only partly confirmed for attention orientation RTs and maintenance percentage of gazing at target half a second after first hit and the optimism robustness hypothesis was rejected for both attention measures. This supports the idea that processes present in optimism bias also play a role in the robustness of optimistic expectancies and their influences on attention in our experiments.
Even though the results of Experiment 1 are promising, one problem with the stimuli used in this experiment is that happy and sad faces could not be assigned to be gain or loss targets differentially across participants. Happy faces always have a positive valence and sad faces always have a negative valence and it would not have been meaningful to tell participants they lose money when seeing a happy face.
These salient stimulus-specific attributes could have differentially influenced attention deployment.
- Hard Changes: A Bouchers World Tale from the Change?
- Rose-Colored Glasses by Jo Ann Simon.
- Lost: One Pair of Rose Colored Glasses?
- Seeing through rose-colored glasses: How optimistic expectancies guide visual attention?
- British Steam: Past and Present.
For instance, in everyday life, we have repeatedly learned that a happy face indicates important emotional information e. Thus, we conducted a second experiment to replicate our effects using non-social and inherently non-emotional stimuli. Experiment 2 was a replication of Experiment 1 with different stimuli.
Spread the word
As both experiments were highly similar, we describe only the details that differ from Experiment 1. If not otherwise indicated, the procedures were identical to Experiment 1. In contrast to Experiment 1, in which the stimuli in the visual search array had to be compared using a rather complex attribute comprising many different features emotional facial expression , the stimuli in Experiment 2 only had to be compared using two clearly separable features color and shape. However, because the emotional face stimuli used in Experiment 1 are highly familiar and overlearned in everyday life, they may generally produce a stronger pop-out effect among neutral distractor faces than the letter stimuli used in Experiment 2.
The procedure was identical to that of Experiment 1 Fig 1 , bottom. The only difference was that letters were presented as stimuli in the visual search task instead of faces. On average, we excluded In addition, outliers deviating more than 3 SD s from the average diameter of a given participant during a particular time interval were eliminated on average 0. Pupil diameter change during the presentation of expectancy cues is shown in Fig 2B. As predicted, the main effect of expectancy cue was significant, F 2. As anticipated, gain and loss cues elicited a larger pupil diameter increase than did ambiguous cues gain vs.
Can a contact lens get lost behind my eye?
Post-hoc pairwise comparisons showed that the differential effect of expectancy condition on pupil diameter change started between 1 and 1. As anticipated, pupil diameter increase was larger for gain and loss cues than for ambiguous cues 0. The RTs are shown in Fig 3B.
- Seeing the World Through Rose-Colored Glasses!
- Joseph Smith's "Magic" Glasses and Other Bizarre Objects from Mormonism;
- 1. Search for the perfect look..
- 2. Visit your optician or an online store.?
- The Inmate Of The Dungeon 1894.
- Smurf-Colored Glasses (episode) | Smurfs Wiki | FANDOM powered by Wikia.
Participants reacted faster to gain targets than to loss targets, showing a main effect of target, F 1. Moreover, participants reacted faster when they expected to gain or lose than when they had ambiguous expectancies, showing a main effect of expectancy, F 2. In accordance with our cue congruency hypothesis, participants reacted faster to gain targets when they expected to gain than when they expected to lose or had ambiguous expectancies gain vs.
Participants reacted faster to loss targets when they expected to lose than when they expected to gain or had ambiguous expectancies loss vs. The last effect is consistent with the idea of an attention bias for positive stimuli. The time to first hit the target results mostly mirror the RT results and are shown in Fig 4B. Participants took less time to first hit gain targets compared with loss targets, showing a main effect of target, F 1. Moreover, they took less time to hit the target when they expected to lose than when they had ambiguous expectancies, showing a main effect of expectancy, F 2.
In line with our cue congruency hypothesis, participants first hit gain targets faster when they expected to gain than when they expected to lose or had ambiguous expectancies gain vs. Furthermore, as anticipated, participants hit loss targets faster when they expected to lose than when they expected to gain or had ambiguous expectancies loss vs. Participants gazed more at gain targets than loss targets in this time span, showing a main effect of target, F 1. As hypothesized by our cue congruency hypothesis, participants gazed more at gain targets within a half second after the first hit when they expected to gain than when they expected to lose or had ambiguous expectancies gain vs.
Moreover, participants gazed more at loss targets within a half second after the first hit when they expected to lose than when they expected to gain or had ambiguous expectancies loss vs.
As hypothesized, a larger pupil diameter increase was evoked by gain and loss cues than by ambiguous cues in Experiment 2. This indicates that gain and loss cues elicited an affective response in our participants that can be attributed to the induction of optimistic and pessimistic expectancies, whereas ambiguous cues did not manipulation check.
Thus, differential effects of attention in our experiment can be attributed to the induction of optimistic and pessimistic expectancies.
Seeing the World Through Rose-Colored Glasses
In accordance with our predictions, the expectancies in Experiment 2 modulated attention deployment, as apparent in the RT and eye gaze data. Participants reacted faster to gain and loss targets when congruent expectancies were induced compared with incongruent expectancies. Furthermore, optimistic and pessimistic expectancies shortened RTs to congruent targets compared with incongruent targets attention orientation. In the eye gaze data, the same effects were observed for the time to first hit the target attention orientation and the percentage of looking at the target half a second after the first hit attention maintenance.
In line with the idea of a general attention bias to positive stimuli, participants payed more attention to gain compared with loss targets when ambiguous expectancies were induced. This attention bias could be explained by a natural Pavlovian tendency to approach reward stimuli. Research has shown that approaching i.
Therefore, a Pavlovian facilitation to approach reward could make people pay more attention to gain compared with loss targets when having ambiguous expectancies. The described Pavlovian tendency to approach reward but not punishment information could also represent an underlying mechanism of this optimism robustness effect because it explains why it might be more difficult to pay attention to loss targets when expecting to gain than to gain targets when expecting to lose.
In conclusion, our cue and target congruency hypotheses and our optimism robustness hypothesis were confirmed for both attention orientation RTs, time to first hit and attention maintenance percentage of gazing at the target half a second after the first hit. Affective states that can be attributed to optimistic and pessimistic expectancies were successfully induced in the experiments reported here. Both experiments demonstrate that optimistic expectancies guide attention toward positive compared with negative stimuli.
This was revealed in the RTs and eye gaze behavior during the visual search task in Experiment 1 for emotional face stimuli except for the time to first hit and in Experiment 2 for non-social letter stimuli. Moreover, in Experiment 2 we clearly demonstrated that pessimistic expectancies guide attention toward negative compared with positive stimuli. As predicted, optimistic expectancies had a stronger influence on attention deployment than pessimistic expectancies—shown by small-to-medium effects in the RT analyses of both experiments and the eye tracking analyses of Experiment 2.
Moreover, this stronger influence of optimistic than pessimistic expectancies on attention was positively associated with individual differences in self-reported comparative optimism bias [ 2 ].
Modulation of attention by expectancy cues is in line with predictive coding theory, which states that humans create a mental template while expecting certain outcomes in their future and compare sensory information with this template [ 26 , 27 ]. Furthermore, our findings correspond to empirical work on the interplay between expectancies and attention deployment to neutral stimuli [ 23 — 25 ].
Therefore, inducing state optimism in the beginning of an experiment or inducing optimistic expectancies through cues on a trial-to-trial basis successfully bias subsequent attention deployment.
Notably, Peters and colleagues could only show rather weak effects of state optimism on attention in post-hoc analyses on alternatively created experimental groups whereas we demonstrated much stronger effects of optimistic expectancies on attention to reward and replicated the effects using non-social stimuli. In addition to replicating results that show optimistic expectancies guide attention toward reward in contrast to punishment, in our second experiment, we demonstrated that pessimistic expectancies guide attention toward stimuli signaling punishment in contrast to stimuli signaling reward.
Notably, this effect was only present when non-social letter stimuli were used. This finding initially arose in Experiment 2 which generally led to stronger effects , which appears to be counterintuitive as social face stimuli would better represent real life situations in which expectancies rely on information with an intrinsic affective meaning. Participants in Experiment 2 reported expectancy cues to be more helpful and important for the subsequent visual search task than participants in Experiment 1 see S1 Analysis.
It is conceivable that the search task in Experiment 2 was simply more difficult because letter target stimuli stood out less among distractors than the face stimuli did in Experiment 1. Therefore, participants probably had to rely more strongly on the information given during the expectancy phase of the experiment.
However, in some conditions, the RTs in Experiment 1 were longer than those in Experiment 2. This implies that participants could withdraw attention from letters more easily than from faces. Because different participants were included in Experiments 1 and 2, it is difficult to draw final conclusions in this respect. Whereas in Experiment 1 only optimistic expectancies influenced subsequent attention to rewarding and punishing stimuli, in Experiment 2 both optimistic and pessimistic expectancies influenced attention but the effect was stronger for optimistic than for pessimistic expectancies optimism robustness hypothesis.