Affect and cognition

Intro

Most cognitive psychologists ignore the issue of the effects of emotion on cognition by trying to ensure that all their participants are in relatively neutral emotional states
Affect = valence feeling state, includes subjective reaction:
-Preferences: pleasant/unpleasant
-Moods: less specific--> "i feel good right now"
-Emotions: complex assortment beyond pleasant/unpleasant (Intense feelings with physical manifestations, and physiological arousal)
Preferences and moods are stable, emotions are short lived
Affect influences thoughts, memories, beliefs and choices, and people vary in their sensitivity to affective influences

People have a slight +ve bias. The effects of +ve mood are more predictable, consistent and interpretable than the effects of -ve mood

+ve more common?
-ve = bigger change from baseline
-ve signal threat, are more varied and are aversive (try to manage feelings more)

Performance and memory

Performance is impaired by high levels of state anxiety

Yerkes-Dodson Law: performance is optimal with a ‘medium’ level of arousal. The ‘optimum’ level lower for hard tasks.
Performance peaks at lower levels of arousal for difficult tasks, and at higher levels for easy/well-learned tasks
Interference?? worry and self-preoccupation interfere with performance, or worry reduced processing efficacy

Losses from failure to gain e.g. choking under pressure:

Its thought that increase in financial incentive = greater performance
Larger incentives increase motivation = elicits improved output and performance.
BUT, when executing skilled tasks, performance increases as the level of incentive increases only up to a point, after which greater incentives become detrimental to performance (Ariely et al. 2009) --> choke under pressure
Chib et al. (2012) fMRI and novel, highly-skilled motor task where win from 6 levels of incentives ($0 to $100). If don't succeed at task, says you lost even though you don't actually lose money.
-found most p's reach peak performance at an incentive level less than $100
-during incentive presentation, BOLD signal in ventral striatum increased with incentive magnitude
-during motor task, BOLD signal decreased with incentive magnitude
=Deactivation of ventral striatum during motor action reflecting evaluation of potential loss (of presumed gain) that would arise from failure to be successful? Larger incentives might be framed as larger potential losses
related to loss aversion. People who are more loss averse = more likely to choke under pressure

Helping and persuasion

Judgment and decision making

Well-being

Good moods = help others in expt. settings:

Baron (1997) in shopping mall, people sig. more likely to help a confederate (by picking up a pen/giving change) when these helping opportunities took place in the presence of pleasant ambient odors (e.g., baking cookies, roasting coffee) than in the absence of such odors. P's also reported sig. higher levels of +ve affect in presence of these odours.

Prosocial behaviour depends on:

Attention: focus on good mood from own good fortune leads to helping, but focus on good mood from others good fortune leads to not helping
Emphasis: reward emphasis = help, guilt emphasis = not help -
Mood maintenance: will helping support mood? Help when the rewards of helping are clear

Present mood shapes memory for past experiences:

more easily remember info thats valence fits current mood state --> recall +ve material in +ve moods, -ve material in -ve moods.
Mood congruity effects occur when episodic memory is better for emotional material that matches a person’s mood e.g. a person remembering more happy than sad events in a story when feeling happy but not when feeling sad.
demonstrated extensively in adults in both normal and clinical samples.
Christodoulou & Burke (2016) shown in 3 & 4 yr old children: chopstick or a popsicle stick in their mouths to either produce or inhibit a smile while they listened to a story featuring happy events for a happy character and sad events for a sad character. Mood ratings taken before and after mood induction showed more happy for smile condition, no change for no smile condition --> found children in the smile condition, but not in the no smile condition, remembered more about the happy character than the sad character in the story
regardless of whether the mood is naturally occurring or experimentally induced (Direnfeld & Roberts, 2006)

Varner and Ellis (1998) 2 expts to elucidate whether mood congruence was result of changes in cognitive activity or physiological arousal. P's presented with words, some relating to depression and others relating to essay writing skills, and subsequently tested on a recall task. 4 induction conditions: depressed mood, neutral mood, arousal (participants engaged in physical activity), and schema (reading statements linked to essay writing).
Mood congruity was found in the depressed induction condition but not in then arousal induction condition, emphasising that cognitive activity, rather than arousal, drives this phenomenon

Number of studies failed to replicate this effect

Rusting and DeHart (2000) studies which did not demonstrate a mood congruity effect usually occur with negative affect --> due to individuals attempting to correct their mood, and revert back to positive affect.
Expt to test: some p's encouraged to try to alter their negative mood. Found that the observed effects rely on subjects’ mood-regulation strategies: those who continually focused on their current affect demonstrated mood congruity, however mood incongruity occurred for those who were encouraged to regulate their emotional state.

Mood improves memory?

Yang et al. (2013) looked at effects of +ve affect on WM and STM- reason that WM involves both storage and controlled processing and STM primarily involves storage processing = if positive affect facilitates controlled processing, it should improve WM more than STM. Task where mood induced by given p's gift of small bag of hard candies. Then did a word-span (STM) and operation-span (WM) task
-mild +ve affect compared to neutral sig. improved WM performance, influence on STM was weaker, and this was not attributable to motivational processes (no differences in spontaneous use of strategies or in time spent of the two tasks) = role of controlled processing??
Curci et al. (2013) demonstrated that WM deficits arise in response to negative affect, and that this is triggered by increases in rumination. Moreover, the emotional valence of the material acted as a mediator in supporting rumination, and consequently, the effects on WM.

Mood state dependent memory = memory is best when mood at retrieval matches that at time of learning

supported by Tulving (1972) encoding specificity principle: retrieval is best when info at encoding and retrieval match
Stronger effects when learning material had personal relevance → according to Bower (1992) causal belongingness hypothesis, memory mainly affected by mood state when p’s believe that their emotional state at learning is caused by the to be learned info as leads to an effective association = likely to occur with personal events
Kenealy (1997) p's examined a particular route on a map, until their learning performance exceeded 80%. The following day they completed free recall and cued recall tests. Strong mood state dependent effects in free recall but not in cued recall = mood state affects memory even when learning is controlled but does so mainly when no other powerful retrieval cues are available
Increases in intensity of mood cause increases in activation of associated nodes in the network. Rinck et al. (1992) considered emotional intensity of the stimulus material – p’s in happy or sad mood rated words for pleasant-unpleasantness. Mood congruency effect for the intensely emotional words, but not the weakly emotional ones on an unexpected recall test

Tendency for individuals experiencing positive affect to judge people, objects and events as more positive than would individuals in a neutral or negative affect state

e.g. Forgas and Gordon (1987) found that those in a positive affect state made more positive judgments and impressions than did individuals in a negative affect state
Positive mood = greater reliance on stereotypes?
Wang et al. (2015) found that p's in a positive mood made more spontaneous trait inferences than those in a negative mood.
-Links to Isen et al (1985): individuals in a positive affect state are more likely to group varied items into the same category.
BUT- Bodenhausen et al. (1994) found individuals experiencing positive affect state are able to avoid influence of stereotypes when making judgments in situations where experiential factors provide motivation to not use stereotypes.

Spachtholz et al. (2016) illustrated the differential effects that positive and negative affects have on memory processes. Whilst positive mood enhanced richness of details during the process of memory consolidation, negative mood increased the strength of the memory. This is in line with research showing that persons suffering from depression tend to have an over-general memory (Brewin et al.1996).

Negative emotions and judgments:

Fischhoff et al (2005) found that whilst fear increased risk estimates, the opposite pattern emerged in response to anger --> these effects extended to past and future events.
Also linked to stereotypes:
-Baron et al. (1990) found that experiencing anxiety generated the illusory correlation effect, wherein one perceives relationships between variables that do not actually exist – a characteristic of stereotyping.
But -ve emotions more complex/produce less predictable findings e.g.
-Jackson et al. (2010) research on specific emotions, rather than +ve and -ve affect, and their effects on stereotyping behaviour - anger, sadness, gratefulness and happiness influence causal attributions --> angry p's attributed the cause of negative behaviour to the target more than sad subjects. Grateful p's attributed cause of positive behaviour to the target more than happy participants, whilst happy subjects took part in stereotyping behaviour more than grateful subjects = effects of mood on judgment depended not only on the valence of the affect, but also on the cognitive appraisal mechanisms associated with the emotions.

+ve mood and decisions:

faster at making decisions, work quickly at simple tasks and will take more risks if the possible losses are minimal, whilst being more cautious when losses are large (Isen et al. 1988).
positive affect is often taken as information to indicate that all is well (Bless and Bohner, 1991), producing suboptimal decisions
evidenced by research on eyewitness testimonies, which demonstrates that subjects in a positive mood are more susceptible to misinformation (Forgas et al. 2005).

-ve affect has different effect on misinformation:

Thorley et al. (2016) compared the impact of negative and neutral moods during encoding and/or retrieval upon eyewitness recall of a non-emotive event. A negative mood during encoding had no impact upon the number of correct details recalled (provided participants were in a neutral mood at retrieval) but a negative mood during retrieval impaired the number of correct details recalled (provided participants were in a neutral mood at encoding). A negative mood at both time points enhanced the number of correct details recalled, demonstrating a mood-dependent memory enhancement. = it facilitates accurate memory recollection, if the negative mood was present at the time of encoding and retrieval.
in line with research finding that negative affect promotes deliberation over decisions (Lambert, et al. 1997) and makes individuals focus on details rather than the big picture (Gasper & Clore, 2002).

Cheerful people are more compliant with persuasive communications, whereas angry or uncomfortable people are less compliant?

Albarracín et al. (2003) levels of processing mediate the effects of mood on persuasive appeal --> moderate processing = affect served as information, high processing = affect deemed irrelevant, low processing = affect not identified
Bohner and Weinerth (2001) found that negative affect facilitates processing only when the persuasive appeal is considered legitimate: such appeals triggered potential threat amongst their participants, triggering deliberation and scrutiny over subsequent decisions.
Raghunathan and Trope (2002), ways in which mood is used as a resource in processing persuasive messages relating to the negative consequences of caffeine consumption - found that induction of positive mood enhanced the impact of negative information about caffeine on high (compared with low; i.e. self relevant persuasive message) caffeine consumers’ attitudes and intentions toward caffeine consumption

Being happy improves health, friendships, marriage, income and performance (Lyubomirksy et al. 2005)

Attentional bias (selective attention to threat-related stimuli presented at the same time as neutral stimuli):

Anxious individuals show on emotional stroop task --> Calvo & Avero (2005) reviewed studies found attentional bias associated with high anxiety in 58% of the studies and findings not significant in remaining 42%
More evidence for attentional bias at early than late stages of attentional allocation: same study found attentional bias towards harm pictures over the first 500ms after stimulus presentation but became attentional avoidance 1500-3000ms after stimulus presentation
Less evidence for depression: Mogg and Bradley (2005) – when attentional bias found in depressed patients it seems to occur for self-relevant negative info which is presented under conditions that allow/encourage elaborative processing
What is the direction of the relationship? MacLeod et al. (2002) trained healthy p’s to develop attentional bias (dot task, target either always in presence of threatening word or in non-threat location). Exposed to stressful anagram task and those with attentional bias showed more anxiety than in other group = attentional bias --> anxiety?

Interpretive bias (tendency to interpret ambiguous stimuli and situations in a threatening way) --> generally agreed that anxious and depressed p's have this:

Eysenck et al. (1991) assessed response bias using ambiguous sentences. Patients with generalised anxiety disorder more likely than healthy controls to interpret such sentences in a threatening way, and there were no group differences in response bias.
Causal?? Wilson et al. (2006) homographs with threatening/neutral meaning – one group trained to focus on threatening interpretations, the other on neutral interpretations. After both shown video of near-fatal accidents. Threatening group became more anxious than other groups. = having interpretive bias --> increased anxiety
Lewinsohn et al. (2001) Measured dysfunctional attitudes in adolescents without major depressive disorder at the start of the study. Year later assessed negative life events experienced over the past 12 months. Those who experienced many negative life events had an increased likelihood of developing major depressive disorder only if they were initially high in dysfunctional attitudes or interpretive bias – because dysfunctional attitudes before onset of depression = a risk factor when exposed to life events
Problems: much more focus on anxiety than depression, and the experiments are artificial, so unclear whether induced biases are the same as those occurring naturally in patients

Memory bias - Williams et al. (1997):

Explicit memory bias (tendency to retrieve mostly negative/unpleasant rather than positive/neutral info on a test of memory involving conscious recollection) typically demonstrated by D p's
Implicit memory bias (tendency to exhibit better performance for negative/threatening than for neutral or positive info on a memory test not involving conscious recollection) typically demonstrated A p's
Because D facilitates conceptual processing and A facilitates perceptual processing of threat related stimuli