What relationships exist between affect and cognition? - Chapter 6

Why is affect in psychology and neuroscience returned?

Long before psychology existed as a science, there was already speculation about the relationship between affect and cognition. Plato stated that emotions (or passions, in his words ) disturbed cognition (or reason ). During the enlightenment, a distinction was made for the first time between cognition, affect and will. Kant again made a distinction between pure reason (intellect, cognition), practical reason (will and action) and judgement (related to pleasure and pain).

These three aspects have been studied separately until quite recently, although together they form one psychological experience. An exception to this was the New Look approach, where much emphasis was placed on the possible influence of affective state and motivation on the fundamental mechanisms of perception and attention.

Only around 1980 did research into the relationship between affect and cognition return to the mainstream of science. Around that time, cognitive psychologists also became interested in how affective state could influence cognitive processes, such as memory and learning. In addition, neuroscience emerged, showing that the neural structures underlying affect are important components in coordinating adaptive cognitive responses. In other words, the realization grew that affect and cognition were closely integrated and that in order to understand the one, a good understanding of the other was also required.

How do affect and cognition interact?

Emotion-focused research focuses primarily on cognitive antecedents and emotion appraisal strategies. Conversely, research into moods rather looks at the consequences of mood on cognitive processes such as attention, memory and social judgments. These areas of research had little to do with each other until it was discovered that the neural basis for both processes is strongly linked.

The Zajonc-Lazarus debate

Zajonc found that affect was not always independent of cognition, but that the two were coordinated by separate processes, which made it possible for them to operate independently. His hypothesis was that the affective properties of a stimulus were identified within a few milliseconds, before conscious cognitive processing. The more exposure effect confirms this; this means that people are more positive about a stimulus as they perceive it more often, without being aware of it. According to Zajonc, therefore, no cognition is needed to cause an affective response.

Lazarus disagreed and believed that an emotion could not arise without a prior cognitive appraisal. The preference that can be seen in the more exposure effect is very similar to a simple evaluation of valence, as discussed in the previous chapter. He also stated that a preference is not the same as an affective reaction: having a preference for something does not mean that you also have a strong affective reaction for this object. He also showed that cognitive appraisals are crucial for determining the type and intensity of the affective response.

Nowadays, researchers generally believe that cognition can take place without consciousness. As a result, they would no longer interpret the fact that preferences do not require conscious processing as evidence that affect is processed independently of cognition. Now, affect is often seen as an integrated part of cognition, where it exerts influence on information processing in different domains (attention, memory, reasoning, making decisions). Emotional responses can determine how a situation is interpreted.

What are different cognitive models of affect?

Appraisal-based models state that the experience of emotion is determined by the way in which a situation is interpreted. However, many of these models do not provide a good explanation for the fact that different emotions can successfully arise quickly without conscious perception of the stimulus. In response, process-based models have now been developed. These state that appraisals can occur at different levels, ranging from low-level action tendencies to high-level conscious decisions. However, much research still uses self-reporting, making it difficult to investigate unconscious appraisal.

When we look at neuroscience, the amygdala can be used as a relevance detector, a neural structure that quickly and automatically detects the relevance of a stimulus. For example, evidence has been found that a positive interpretation of a situation causes activation in the vmPFC, while a negative interpretation causes activation in the left ventral amygdala. Activation in these parts is therefore directly related to differences in appraisal.

Information-processing approaches have been developed separately from appraisal-based models. They focus in particular on behavioral methods designed to determine how people orientate towards an event or object and how they remember it. However, this did not look at the role of affective state. Interestingly, this approach could not explain how biases arise in appraisal. Appraisal theories use 'appraisal' as the explanation for emotions, but information-processing models attempt to understand why appraisal itself is often biased. 

Applying information-processing models to emotion and emotional disorders ensures that biases can be viewed at an unconscious level. This gives this method a major advantage over self-reporting methods. Furthermore, the method can easily be used in combination with brain imaging techniques such as Event Related Potential (ERP). These models were used for the first time around 1980 to investigate emotion and emotion disorders. Affect ensures that the cognitive system receives a representation of the value or significance of a stimulus, enabling rapid actions and responses to environmental requirements.

How is affective information processed?: Empirical data

A mood or emotion can cause a cognitive bias, making mood-relevant material preferred by the processing system. This bias also influences mood and emotions. The cycle is further influenced by individual differences and external factors. Our views on the world are largely determined by fundamental cognitive processes: perception, attention, interpretation, assessment and decisions, and memory. A bias in these processes has a strong influence on one's view of life.

Which role do perception and attention play in processing affective information?

Information-processing approaches assume that there are limits to how much information can be processed at the same time. This leads to a high degree of selectivity in attention. So there are mechanisms that separate important information from the flow of irrelevant information. Attention must, however, be extremely flexible, because irrelevant information may become relevant at the next moment.

There is a debate between early versus late selection theories. This is concerned with whether information is identified before or after attention is focused on a particular object. It seems logical that part of the processing of affective significance already takes place before perceptual processes are complete, to explain the effectiveness and speed of attention.

An object can be affectively significant because it has been 'hard-wired' in our system from birth. Many stimuli, however, are because we have learned that they are important and because we have formed selective associations between objects and emotional responses. Based on Zajonc's theories, we can state that a distinction between good and bad stimuli (valence) is made very early in the attention process. This creates an attentional bias, especially for impending stimuli. This bias can occur at an implicit (unconscious) level and exists at every stage of processing, but bias towards negative objects generally occurs very early in the process.

Affective processes can influence perceptual processes both directly and indirectly (via attention mechanisms). This is because attention to a specific object or location can activate parts of the visual cortex that deal with the initial registration of stimuli, which will improve the perception of the object.

What results from behavioral studies?

Behavioral tasks have been developed to look at the extent to which people are biased towards certain types of stimuli. Search tasks, for example, show that people find a target more quickly when it is distinguished by a single property (for example, color) than when it is distinguished by a composition of properties (color, shape, etc.). The size of the display (and therefore the number of distractors) does influence the search time for these compound searches, but not the single one. This is because a serial search strategy must be applied, while a single property such as color stands out immediately.

It turns out that when people need to detect faces, they find angry faces faster than happy faces. Attention is therefore drawn to these impending stimuli. This is called the threat superiority effect. The size of the display had less effect on reaction time with the angry faces than with the happy faces. This indicates that angry expressions are detected more efficiently. The same effect was not found for sad faces, indicating that the effect is caused by threat and not by negative valence. The effect has also been found for many other objects, some of which are of evolutionary importance (threatening faces, snakes, etc.), while others are learned (guns, injection needles, etc.).

A Stroop task is also often used for attention research. People get to see words, but they must ignore their meaning and only mention the color in which the word is printed. However, the meaning of the words does cause an interference effect: it disrupts the processing of the colors. It appears that people take longer to name the color with negative words than with positive words. This indicates a general vigilance for negative social information compared to positive social information.

It has also been found that the affective significance of a stimulus can result in improved perception. In other words, we see or hear better. Phelps demonstrated this in her research where people were better able to distinguish contrast after seeing a frightened face than after seeing a neutral face. This effect is strongest when it comes to focused attention (i.e. on one stimulus), compared to spread attention (ie on multiple stimuli spread across the image). These findings provide a strong indication that negative stimuli can have both a direct and an indirect effect (via the impact of attention) on perception, as discussed above.

What results from brain imaging studies?

Multiple neuro-imaging studies have shown that activity in the amygdala correlates with more activity of neurons in the extrastriate cortex when emotional stimuli are observed. This is in line with the hypothesis that emotional stimuli can enhance sensory processing, especially when it comes to threatening stimuli.

It was also found that faces are mainly processed in the fusiform cortex. Research showed that the presence of a frightened facial expression alone led to an increase in activity in the fusiform gyrus even when no attention was paid to these faces. This was not the case when the faces had a neutral expression. The same effect was found for activity in the amygdala. The consequence of this greater activity in the sensory cortex after seeing an impending stimulus is that dangerous stimuli are noticed earlier than other, less salient stimuli. This effect has been confirmed several times by research with patients with left spatial neglect (they 'miss' stimuli in the left visual field, but these are unknowingly processed).

A similar effect has also been found for auditory stimuli, which involves activity in the superior temporal sulcus (STS). So for both visual and auditory modality, the presence of threatening information alone can lead to additional stimulation of sensory processing.

The amygdala probably plays a crucial role in regulating the enhanced sensory response to impending stimuli. More specifically, the presence of a negative stimulus stimulates cells in the amygdala, which in turn causes an increase in cell activation in the sensory cortex. The amygdala is perfectly positioned for this role, as it receives sensory information from every modality, and sends projections to many subcortical and cortical areas. As a result, the amygdala can influence (and be influenced by) multiple perceptual and cognitive processes.

How much time does affective processing take?

It may be that in an attention-seeking task, attention is directed just as quickly to positive as to negative stimuli, but that responses to negative stimuli are selected more quickly. fMRI and PET data cannot properly show this, because they do not give a precise representation of the passage of time. This effect seems logical in any case, because in the presence of danger it is useful to be able to act quickly.

ERP (event related potential) can produce a good time display. Data from ERP research that looks at a component called P100 (which shows changes in attention allocation), indicates that there was indeed a stronger attention for negative stimuli than for neutral or positive stimuli. So, maybe differences in response selection played a minor role in behavioral studies, but the ERP data indicates that the results cannot be entirely caused by longer or shorter response selection. Later it also became clear that, regardless of whether a response has to be given, attention is immediately drawn towards fearful faces / stimuli, compared to happy faces.

Another ERP component, called C1, occurs 60-90 msec after the stimulus appears. The peak in C1 is enhanced by aversive stimuli. This effect also remained intact when it came to neutral stimuli that were conditioned to have had a negative association. Due to the extremely short time between stimulus presentation and C1 response, it seems likely that sensory processes are immediately enhanced for aversive stimuli. This is in line with the idea that threat assessments take place very quickly.

Judgment and decision-making

Since affect can have strong attention bias effects on neural mechanisms, it seems likely that the same affective properties of the environment also have a profound effect on how we assess and evaluate our environment. Important decisions in life are almost always associated with affective significance. This idea stems from research into people with brain damage, who as a result no longer received certain affective signals and as a result were unable to make certain basic choices. In addition, general mood has an important influence on how we evaluate social situations and our own satisfaction with life.

Is affective processing necessary for effective decision-making?

Physical state, or somatic state (for example, a ' gut feeling') can arise from primary inducers (learned stimuli that evoke positive or negative emotional states) or secondary inducers (a reminder of a real or hypothetical primary inducer, which can cause a physical state when it comes into the working memory).

Primary inducers directly activate the amygdala, which then causes a physical condition caused by the release of various neurotransmitters into the brain stem. Secondary inducers, on the other hand, activate physical states through cortical circuits, in which the vmPFC plays an important role. This can combine knowledge from secondary inducers into physical state patterns that reflect what it feels like to be in a certain situation.

The 'as if body loop' describes that the activation of representations of physical state in the brain voice (by a secondary inducer) can result in changes in neurotransmitter release, without actually involving the body. However, these are felt as an actual physical condition.

Patients with bilateral damage to the PFC have major problems making decisions in daily life and often make decisions with disastrous consequences. This is probably because they lack a somatic marker (this is the ' gut feeling'). The somatic marker hypothesis (SMH) is designed to explain this. The problems with decisions are caused by a deficit in the emotional mechanism that rapidly signals the possible consequences of an action. The gut feeling is crucial here. When this somatic marker works well, people are led away from bad decisions, towards an advantageous response.

Criticism of the theory points to the possibility that a somatic marker is more a consequence of making a risky decision, rather than a signal that is used during the decision.

Does mood influence personal and social judgments?

Many studies have shown that mood can have a strong effect on our judgments about objects and people. This can be explained by the fact that people make evaluations that are consistent with their current mood, so when they are in a positive mood, their judgments are also more positive. When a certain mood is activated, all events associated with that mood are also triggered in our memory.

For example, people in a negative mood think that certain negative events (getting a disease, for example) are more likely than people in a neutral mood. Even when it comes to judging oneself, mood has a major impact on people. Happy people give themselves a higher score for self-confidence than people with a negative or neutral mood. In addition, they attribute more positive traits to themselves. This effect is diminished when a question is asked about which people have to think deeply.

The affect-as-information approach

The affect-as-information approach argues that people make often judgments by asking themselves 'How do I feel about it?'. So the idea here is that people use their current affective state as information that helps to form a judgment or evaluation. Thus, unlike many other approaches that view feelings as output, this approach sees feeling as an input for cognition and behavior. There is even some evidence for a neural network that promotes this relationship.

However, there are exceptions. If the feelings are not attributed to the object being judged, then the relationship disappears. So, if the object is clearly not the cause of the vote, it cannot serve as input.