Thinking Clearly about the Role of Emotions

How Do You Feel? Introducing Emotions

You may think that you know exactly what emotion is, but as a scientist you need to take a step back and question any easy assumptions.

Consider this example: in Star Trek IV, the computer asks Mr Spock, ‘How do you feel?’ This question stumps the friendly Vulcan because he has no emotions and is always confused by his ‘flawed feeling human’ friends. If, suddenly, he felt an emotion, how would Spock know what it was? That’s a tricky question! You know when you feel happy, but what is happiness? The answer to that question goes well beyond the scope of cognitive psychology, or even psychology (try philosophy!).

In this section, we describe what cognitive psychologists mean when they refer to emotion, show how people learn emotions and explore the thought processes surrounding emotions.

Emotions, moods and affective states are similar but distinct concepts:

Emotions: Brief, but intense, experiences
Mood: Prolonged, low-intensity experiences
Affect: The whole lot of experiences, including emotions, mood, preferences and arousal (the bodily reactions to stimuli such as sweaty palms or increased heart rate)
Valence: The subjective quality of a particular event, such as thinking something is good or not good

Looking at ways of defining emotion

Broadly speaking, psychologists describe emotions as a combination of the following:

Physiological changes and arousal: How your body changes due to a stimulus. For example, when you’re afraid, your heart rate increases and your pupils dilate.
Facial expressions: When you feel a particular emotion, your face contorts into a specific pattern. For example, when you’re happy, the corners of your mouth tilt up, and if you’re very happy, you open your mouth to reveal your teeth – in other words, you smile!
Cognitive appraisals: When you feel an emotion, you evaluate the event cognitively and decide whether it’s positive or negative. For example, if you’ve just been asked out by the guy you’ve fancied for months, you’re probably able to work out that you should be happy.

At this point, you may be thinking that people display emotions and appraise their feelings in different ways. Indeed, a number of experts believe that cross-cultural differences exist in facial expressions – if you meet someone from a tribal village in Papua New Guinea and he reveals his teeth, you may not know for certain whether he’s happy or hungry.

American psychologist Paul Ekman claims that six basic emotions and their associated expressions are universal. They’re the same across all peoples of the world, because they’re genetic: happiness, surprise, anger, sadness, fear and disgust. You can classify these six emotions by relating them to the fight-or-flight response as follows:

Approach emotions: Happiness, surprise and anger make you feel as though you want to draw nearer to the thing causing the emotion.
Withdrawal emotions: Sadness, fear and disgust make you feel as though you want to move away from the thing causing the emotion.

You can also use the response of the sympathetic nervous system (part of the nervous system that controls the basic physiological responses to stimuli) to classify these emotions. For example, happiness, anger, surprise and fear result in an increase in activity, whereas sadness causes a depletion of the emotional resources.

Ouch! Developing emotional responses

Emotions are externally mysterious. Why does something become emotional for you? Why do some people develop phobias? Places, items, songs and people all have emotional value – some more than others, depending on the person and his experience.

One of your authors has a phobia of elastic bands – we don’t want to stretch the point (groan!) but he gets anxious when he sees one. On the other hand, driving down the street where his first love lived always makes him feel happy. How do such connections – between, say, a street and an emotion – develop?

One theory to explain how people discover that particular things have emotional value is through classical conditioning. Simply put, emotional value can be paired with an object. People learn to associate a particular place or thing with a particular feeling or emotion.

Ivan Pavlov was one of the first scientists to test this phenomenon systematically, having noticed that the dogs in his lab salivated at the sound of a trolley bringing the food, instead of just the sight of the food. The dogs seemed to have paired the sound with the food they’d receive.

Similarly, classical conditioning has been used to explain how some phobias may develop. Fear conditioning is when a particular neural stimulus (such as a word) is paired with an electric shock. The unpleasant electric shock causes arousal and fear, as shown by elevated skin galvanic response (slightly sweatier hands). Eventually participants associate the stimulus with the shock and the stimulus alone produces arousal.

Liz Phelps, an American neuroscientist, identified that the amygdala (part of the brain about 5 centimetres behind each eye) is critically important in fear conditioning. She tested a patient, SP, who had damage to the amygdala. SP was able to remember an experiment and report receiving electrical shocks in it. However, she never showed the elevated skin response associated with fear of the stimulus. Thus, Phelps concluded that the amygdala was critical for processing fear.

Although fear conditioning is readily demonstrable – you can pair almost any previously neutral stimulus with a fear response – some stimuli can be fear-conditioned much more easily. Snakes, spiders and other normal phobia objects (unlike a fear of elastic bands) can be fear-conditioned more quickly and more easily than less terrifying stimuli (such as a cute guinea pig). So humans may have some kind of genetic predisposition to fear certain animals, and only minimal learning is required to develop this fear (although check out the nearby sidebar ‘Watch and be afraid … very afraid’). Alternatively, these fears may be more socially and culturally expected so that people expect to fear them more.

Advertisers use a similar form of emotional classical conditioning (evaluative rather than fear-based) when they pair a product with something desirable. The old-fashioned, sexist example is pairing scantily clad women with fast cars. Advertisers also use endorsements from (supposedly) popular celebrities so that you pair the feeling of liking the celebrity with the product. Advertisers can also make their adverts funny so that you connect the amusement and liking of the advert with the product.

People can also learn fear, and by extrapolation all forms of conditioning, through observation. You don’t need to receive the shock to be afraid of a particular object; simply witnessing someone else getting the shock is sufficient for you to pair the stimulus with the shock.

Thinking about emotion

Here’s an interesting question that fascinates cognitive psychologists. When you have the physiological responses that are generally associated with an emotion, do you need to know why they’re happening in order to feel the emotion? Put another way, can you have an emotion without any cognitive awareness?

On one side of the debate, the American-Polish social psychologist Robert Zajonc claims that you can experience emotions without any conscious awareness as to why you have them: this is called the affective primacy hypothesis. He suggests that affect (the collection of all your emotional experiences) and cognition are based on two separate systems.

As evidence, Zajonc demonstrated the mere exposure effect, where people rate things more favourably if they’ve seen them before, irrespective of whether they remember seeing the object.

In a contrasting view, Richard Lazarus, an American psychologist, proposes that cognitive appraisal plays an important role: you can’t have an emotion without thinking about the object or event. Individuals carry out a primary appraisal of a situation, regarding the situation as positive, negative or irrelevant. They then perform secondary appraisals, to assess their coping ability by implicating someone as responsible and to establish the expectancy of the event occurring in the future. They then re-appraise and monitor these conclusions.

To back up cognitive appraisal, Lazarus presented emotionally stirring films, after which participants were given instructions to intellectualise or deny the events of the films. These instructions reduced the physiological response to the films relative to a control condition, indicating the importance of thinking about events and the emotional response.

So, which view is more accurate? Do you need to think before you have emotion or not? Well, neuroscience provides some answers: noted American neuroscientist Joseph LeDoux identified two emotion-related neural circuits:

A slow-acting circuit: A slow-acting circuit that involves conscious detailed cognitive processing of an emotion, supporting Lazarus.
A faster circuit: A faster circuit that bypasses the conscious processing and the cortex, supporting Zajonc.

Therefore, both are correct! Fast emotional responses don’t require thought whereas slower ones do – which is why you should count to ten when you get angry!

Recognising the Reach of Emotion

Most experimental studies show that sad moods are detrimental when people are performing cognitive tasks. Sad moods impair performance on reasoning, thinking, memory and face recognition, whereas happy moods tend to be beneficial in many of these tasks. But things aren’t quite that simple (they never are in psychology!). These mood differences tend to occur only during difficult or complex tasks. Fewer differences exist between happy and sad people with simpler tasks.

Emotions affect cognition on many levels. In this section, we explain how they impact all the basic processes that we describe in this book, from perception and attention to memory, language and thought.

Attending to emotions and perception

Mood affects how you perceive the world and attend to things in it. Here we describe how this happens.

Emotional perception

On some occasions, emotion can affect what you perceive. When provided with ambiguous stimuli, sad people tend to interpret them negatively, whereas happy people interpret them positively. For example, when participants are shown a neutral face and asked to rate its emotion, sad people perceive it as sad and happy people as happy.

Differences also exist as to where sad and happy people look when viewing visual scenes. During face-perception tasks, sad people tend to look at features other than the eyes, whereas happy people tend to look at the eyes more. Further experiments show that sad people look around a room more than at the person with whom they’re talking.

Emotion and attention

Generally, emotional material grabs your attention more than non-emotional material. This tendency makes sense, because emotions are likely to portray important information for your survival or well-being.

In Chapter 7, we introduce the Stroop effect (where naming the colour of ink in which a colour word is written takes longer when the word’s meaning fails to match the ink’s colour: say, ‘red’ written in green ink).

The similar emotional Stroop task uses emotional or phobia-related words in different colours (for example, ‘snake’ for those with snake phobias). Emotional words interfere with colour ink naming more so than neutral words, showing that emotion interferes with attention.

Researchers have also used the Posner cueing paradigm – where a cue precedes the location of a target on valid trials and doesn’t precede the location of a target on invalid trials (refer to Chapter 7) – to examine how emotion affects attention. When the cue is a threatening stimulus (such as an angry face), the effect of an invalid cue is greater than when the cue is neutral. In other words, participants fixate on an angry stimulus and can’t disengage from it. Emotional cues hold the attention.

But instead of showing simply that emotion affects attention, research also reveals that specific emotions affect attention in different ways. In visual search tasks, people find angry faces easier to detect, even in a crowd of other faces. But the same effect doesn’t occur for happy faces, highlighting how valence (refer to the earlier section ‘How Do You Feel? Introducing Emotions’ for a definition) influences cognition.

Experiments have unearthed another effect in sad and depressed people: defocused attention is where fixing attention on one thing is difficult. Instead of being a spotlight, attention is more spread out. Scientists have long known that depression is associated with many intrusive thoughts. If depressed people can focus their attention on a particular task and block out these intrusive thoughts, they’re better able to perform the task.

In one experiment, sad participants were presented with words surrounded by a coloured frame in the left- or right-hand side of a computer screen. When given a word-recognition task, happy and sad participants performed equally well. For each word they recognised, they were then asked about the coloured frame and the position. Happy participants were unable to remember either bit of extra information, but sad participants did. Sad moods seem to cause people to attend to lots of extra, unwanted, irrelevant information.

Both the emotion of the stimuli and the emotion of the observer affect how something is likely to be attended to. We cover further attentional biases that emotions cause in the later section ‘Worrying about Anxiety’.

Remembering to cover memory and mood!

Mood affects how participants remember information, what information is remembered and what’s recalled. Here we describe these effects.

Mood dependency

Many people remember things more effectively when they’re in the same emotional state as when they learnt them (the encoding specificity principle from Chapter 9). A version of this effect has been discovered using emotions: if people learn items in a particular mood they’re better able to recall that information when in the same mood (known as mood-dependent memory). So, if you revise for an exam when you’re happy, you perform best during it when you’re also feeling happy.

In one study of mood-dependent memory, participants were made to feel a particular mood: in this case, happy or sad (called mood induction). They then learnt a list of words. Half the participants recalled the words in the same mood, whereas the other recalled the words in the contrasting mood. Participants whose mood matched from learning to test recalled more words than those participants whose mood didn’t match.

Research also shows the effects of mood-dependent memory on recalling childhood memories: when happy, people recall about four times more happy childhood memories than sad ones. Clinically depressed patients tend to rate their own parents as having been more rejecting and distant when the patient was a child. But when not depressed, the same patients rate their parents as being warmer and kinder.

Although the logic of mood-dependent memory is simple and compelling, it isn’t found consistently. A number of researchers have failed to find it, leading some people to consider it an unreliable effect. In fact, three principles exist in order to obtain mood-dependent memory effects:

The emotion must be seen to be causally belonging to the material. That is, the information must somehow relate to the emotion.
Mood-dependent memory is more likely for less meaningful stimuli or ambiguous stimuli. That is, mood affects memories that are not very important.
Without any other cues to memory, mood may be useful. Thus, mood-dependent memory effects are more likely in free recall experiments than recognition or cued-recall tests.

Despite not being consistently found, mood-dependent memory effects are more likely if the mood is strong and stable and is used when learning about the information. Just being in a particular mood doesn’t mean that you remember things learnt previously in that mood – you remember them only if they relate to the mood. For example, depressed patients often report intrusive negative thoughts: that is, the negative mood causes negative memories to resurface.

Mood congruency

Mood-congruent memory is similar to mood-dependency and is where your present mood aids the recall of information that’s congruent with the mood, regardless of your mood at the time the information was learnt. Therefore, currently happy people tend to remember happy events more than sad events and currently sad people tend to remember sad things better than happy things.

Mood-congruent memory effects have been found in clinically depressed patients. Depressed people given word lists to memorise tend to recall more negative items than positive. Furthermore, mood-congruency is present in the amount of cognitive effort paid to particular stimuli. Participants in a sad mood are likely to pay more attention and spend more time viewing sad stimuli than participants in a happy mood.

These effects of mood-congruency may be down to mood congruency during learning of the information, mood congruency during the encoding or storage processes, or better mood-congruent recall. Researchers tested this effect by getting participants to learn emotional material after or before a mood induction. Typically, mood-congruency effects are much stronger at learning than at retrieval.

Encoding when emotional

In Chapter 9, we present the levels of processing framework, which identifies when you’re more likely to process things more deeply and so remember them better. The more elaborate and semantic the learning, the more likely you are to remember something. Some research explores how elaboration is affected when you feel emotional.

Using mood induction, experimenters found that sad participants don’t benefit from deeper encoding (at least when limited time is available to process information). The conclusions are that sad mood impairs memory for elaborative encoding and only under higher cognitive load. Other researchers found that people in sad moods tend to find using deeper encoding more challenging than people in happy moods: that is, they don’t show the same levels of cognitive processing effects. Participants in a sad mood, or depressed patients, recall words that they should’ve deeply encoded at the same level as shallowly encoded words. The conclusion is that sad mood disrupts effortful coding.

Organising thoughts when you’re emotional

Sad participants seem unable to organise material to be remembered. For example, when word lists are given to sad, happy and emotionally neutral people, typically the sad people recall fewer words. However, if the lists of words are organised in a meaningful manner (that is, words relating to similar concepts are put together), all participants perform equally in this recall task. When the list of words is highly disorganised, sad participants show an even larger recall deficit.

Although sad people can better recall organised information, they’re more likely to remember associated words falsely when using the DRM paradigm (refer to Chapter 12). Here participants are presented with a series of words related to a particular concept (say, ‘doctor’, ‘nurse’ and ‘medicine’). Sad people are also more likely to make false recognitions when researchers present lure words in a recognition test (words related to the concept, for example, ‘hospital’). The lure word isn’t presented first, and so shouldn’t be recognised, but sad people tend to recall seeing more lure words than neutral people.

Another method for showing organisation in memory is looking at the order in which words are recalled. Typically people recall words in a clustered manner – they tend to recall words related to a similar concept at the same time. Sad participants don’t show this effect. Depressed mood thus affects people’s ability to organise material.

Speaking about language and emotion

Mood also affects how people comprehend information. In one study, participants read a passage that was hard to comprehend without a title providing context. The ambiguous passage describes, say, washing clothes without using the word ‘washing’ or ‘cleaning’ (try it in Chapter 12). Sad participants comprehended and recalled much less than neutral participants and were less confident in judging what they’d remember.

In lexical-decision tasks (where participants have to identify whether a word is a word or not) happy participants are faster when the words are happiness-related than when they’re sadness-related, highlighting how mood affects the very early stages of coding. Some degree of precision applies with these effects, too. Sadness doesn’t speed up decisions on general negative words, though, only specifically sadness-related words.

Another assessment of how mood affects reading comes from word-naming tasks, where participants simply have to read words. Again, research finds mood-congruency in the speed of naming words: happy people read happiness-related words faster than sadness-related words.

Thinking that you may be in a mood

Mood has pervasive effects on how people think. Consider someone who’s feeling sad or depressed. If you say something innocuous, the person often immediately associates it with something bad. This reaction isn’t simply a depressed person moping but a consequence of the semantic and emotional network.

Mood and how you think

Research shows that if you ask depressed people to name a type of weather beginning with the letter ‘s’, they often describe less pleasant weather such as a ‘storm’. If you ask happy people the same thing, they say something like ‘sunny’. Mood thus affects what springs to mind. Try it on your friends to identify the Down Dereks from the Happy Harrys!

Mood also influences preferences and likes. When you’re happy, you tend to prefer to be outside and active; when you’re sad, you prefer to be inside and sedentary. Happy moods also cause you to integrate your knowledge into larger, more inclusive units. People include more things in positive categories when they’re happy.

Emotional decision-making

Mood also affects the way you encode messages, with sad people processing the world very differently to happy ones.

When presenting persuasive messages to happy and sad people, the latter appear less likely to be swayed by a weak argument than happy people. Sad people are more likely to process the information deeply using more elaborative cognitive processing (which may seem contradictory to what we say at the start of this section, but it isn’t, honest!), a case in which sad mood seems to benefit cognitive processing.

Happy people tend to employ faster and simpler cognitive processing, characterised by more heuristic use (mental shortcuts based on schemas and stereotypes; see Chapter 12) and more superficial encoding. They look more at the gist of situations, without focusing on the detail, and tend to use more open, flexible and creative processing.

Sad people’s processing tends to be slower, more systematic and analytical. They focus more on the details of scenes and are more vigilant in their processing and less flexible in their problem-solving. Perhaps they’re deliberately more accurate at these tasks to improve their mood by being more successful.

Looking Behind the Reality: How Mood Interacts with Cognition

A number of theories explain how mood interacts with cognition. In this section, we review briefly four of these models. Most were devised with the intention of explaining cognitive deficits in people with depression.

Activating feelings: Emotional network

In 1981, Gordon H Bower, an American cognitive psychologist, used a neural network model to explain how emotions can affect memory and other cognitive structures. Simply put, knowledge is represented in memory through a series of nodes (refer to Chapter 12). Each node represents an idea or a construct and is connected to other nodes that represent related constructs. Emotional nodes are also connected to the physiological systems. Whenever an emotion is activated, it activates all the nodes connected to it (called spreading activation). When the activation reaches a certain threshold, the ideas become conscious. Critically, a node can be activated by external or internal causes.

We summarise this theory in Figure 20-1. The ovals represent nodes of semantic information, autobiographical memories, physiological responses and behaviours. The pentagons represent the emotion.

© John Wiley & Sons, Inc.

Figure 20-1: Bower’s network theory (shaded items at the bottom are human behaviour).

So, if the emotional node ‘happy’ is activated, it causes the face to smile and the physiological system to release endorphins. It activates the connected memory nodes of a happy time (say, a particular holiday or a successful date) and also any items learnt during a happy state. Activating these nodes causes them to be easier to bring to the conscious mind: that is, the affect (flip to the earlier section ‘How Do You Feel? Introducing Emotions’ for a definition) primes all connected nodes.

This simple model accounts for mood-dependency and mood-congruency effects (which we describe earlier in the section ‘Remembering to cover memory and mood!’). For mood-dependent memory, the idea is that word lists are learnt while attached to the emotional node. If this link is strong, the connection between the word list and the emotion is stronger. Thus, when the emotion is reinstated, the words are as well. Mood-congruency is accounted for because when an emotion is activated, so are all the attached nodes, making them easier to access.

Maintaining focus: Resource allocation model

American psychologists Henry C Ellis and Patricia Ashbrook identified that, based on the network theory (refer to the earlier section, ‘Activating feelings: The emotional network’), participants experiencing a particular emotion are likely to have related thoughts activated. This effect is particularly noticeable in depressed patients with intrusive thoughts. The extra emotional thoughts entering the mind mean that the attentional system (refer to Chapter 7) must work overtime to block out distracting thoughts, which isn’t easy. So, unwanted emotional material takes up too much of the working memory.

This resource allocation model (RAM) is based on the principle of limited attentional and/or working memory resources being available. Emotional states moderate the amount of resources available for other tasks, but take up some of those resources with irrelevant information.

This model suggests that mood-congruent memory effects occur because the emotion causes the attentional system to allocate more resources to mood-congruent stimuli. It explains why mood causes deficits in cognitive processing information but fails fully to describe mood-dependent memory effects.

Trusting your feelings: Informative emotions

The affect-as-information model suggests that when presented with a particular stimulus, people simply decide how they feel about it and this decision guides their cognitions. (We define ‘affect’ in the earlier section ‘How Do You Feel? Introducing Emotions’.) In a memory experiment they see a word and decide whether it ‘feels’ good or not (a very subjective feeling indeed). Subconscious knowledge of the mere exposure effect (which we describe earlier in the section ‘Thinking about emotion’) means that if the word feels good they’re likely to think that they’ve seen it before.

This model requires establishing how you feel at that particular moment and linking it with the stimulus presented. People consult their emotions to make quick and not fully thought judgements about something, suggesting that judgements aren’t based on detailed elaborative coding.

One piece of evidence for this model comes from a telephone survey. Experimenters telephoned a random selection of the public and during the conversation asked for quick judgements about their mood and general life satisfaction. On sunny days, participants tended to report being happier and more satisfied with their lives. On rainy days, they were less happy and less satisfied. But when researchers made the participants aware of the probable source of their mood, by asking them about the weather, these effects disappeared. Thus, people tend to base a judgement about their mood on the weather, unless they know that it’s the cause.

The affect-as-information model accounts for mood-congruent memory effects by suggesting that mood guides the search process used by cognition. Mood guides how people look through their memory system, but only affects cognition in the absence of an identifiable reason for the emotion. If the emotion stems from an unrelated stimulus, it’s ignored.

Although this theory is simple and general, a number of studies find that its basic premise (that only emotions without an obvious cause are used to guide cognitions) is false. The theory also predicts that mood only affects judgements (whether evaluative or recognition-based), and yet plenty of data show that mood impacts learning and encoding. Thus, affect seems only to be used as information under a limited set of conditions when quick and schema-based processing is all that’s required.

Choosing an appropriate processing type: The affect-infusion model

Incorporating the preceding models, Australian psychologist Joseph Forgas developed the affect-infusion model (AIM). This model brings in evidence suggesting that emotions affect cognition in varying ways and that this effect is highly context-dependent. The model is based on the idea that humans put the minimum amount of cognitive effort into performing any task, employing the easiest strategy to succeed at a task. However, different tasks require different strategies and levels of effect. Several types of cognitive processing exist and emotion is infused into the processing for only some of them.

The different types of processing are based on the following:

Features of the information: For example, familiarity, complexity
Features of the person: For example, motivation level, cognitive capacity
Features of the situation: For example, expectations, time pressure

These factors combine and allow for four possible processing types:

Direct access: This simplest type is crystallised (it can’t be altered): you have set-in-stone opinions or facts. You use this form of processing for highly familiar tasks, such as stating your name or where you grew up. These facts are so readily available that you need no elaborate processing, and the processing is so robust that it resists the influence of emotions.
Motivated: When you have a strong desire to search for a particular piece of information. This processing is goal-directed, meaning that you use it to seek out particular information. It involves quite a lot of cognitive resources.
Heuristic: Relies on minimal information and is used when no personal relevance or motivation is present to process more deeply. In other words, it’s a shortcut. For this reason, processing may be based on irrelevant social factors or, indeed, irrelevant internal factors, such as emotion. When using heuristics, people may say, ‘I feel happy; therefore, that thing must be good’.
Substantive: The most elaborate and extensive processing type is used only when easier strategies can’t be. It’s the most susceptible to the influence of emotion, and is employed whenever a task is complex or personally relevant, or when you need accuracy. Mood can affect many different sub-processes within this type, which is why it’s the most likely to be affected by emotions.

Motivational and substantive processing are those employed when more elaborate encoding is required. However, affect is likely to have a bearing on heuristic and substantive processing more than the others.

Many factors can affect which processing a person employs. Table 20-1 summarises a few of these influences and how they may work.

Table 20-1 Factors Affecting the Type of Cognitive Processing Employed within the Affect-Infusion Model

Processing Type	Features	Degree of Affect Infusion
Direct access	Familiar Not relevant Not important	No affect infusion
Motivated	Relevant Important Specific motivation	Low affect infusion
Heuristic	Typical target Simple processing Low cognitive capacity Positive emotional state Accuracy not required	High affect infusion
Substantive	Atypical and unusual target Complex processing High cognitive capacity Negative emotional state High motivation for accuracy	Highest affect infusion