Figure 189

Memory performance (recall and recognition) as a function of learning conditions (generate vs. read) and match vs. mismatch of moods at learning and test. Based on data in Eich and Metcalfe (1989).

were no faster than normals in thinking of negative personal experiences. Presumably it was so painful for the suicide attempters to retrieve unpleasant personal memories that they inhibited the retrieval of such memories. Second, mood-state effects are strongest when participants learn and remember personal events (Ucros, 1989). It is not clear on Bower's (1981) original network theory why this should be so.

Other theoretical perspectives

In the years since Bower (1981) put forward his network theory, there have been various attempts to provide more adequate theoretical accounts of mood and memory. Bower (1992) offered an explanation of why mood-state effects are greatest when personal events are learned and remembered. He argued for a causal belongingness hypothesis. According to this hypothesis, memory is only affected by mood state when participants believe that their emotional state at learning is caused by the to-be-learned stimuli. Causal attribution leads to an effective association between the stimulus and the emotional state. This is much more likely to occur with personal events (e.g., feeling delighted after succeeding in an important examination) than when an emotional state is induced before presenting the learning task.

Eich and Metcalfe (1989) argued that mood state has more effect on internal events such as reasoning or imagination than on events that are more closely determined by external factors. Thus, memory for internal events is more susceptible to mood effects than is memory for external events. They tested this hypothesis in two ways. First, participants who had been put into a happy or sad mood were given either a read task or a generate task. The read task involved reading a category name followed by two exemplars (e.g., precious metals: silver-gold), whereas the generate task required participants to complete the last word (e.g., precious metals: silver-g...?). It was assumed that internal events would be more important with the generate task than with the read task. Second, memory was tested with either free recall or recognition, and it was assumed that free recall would place more demands on internal events than would recognition memory. Thus, it was predicted that mood-state-dependent effects would be greater with the generate condition than with the read condition and with free recall than with recognition memory.

The main findings were in line with the predictions (see Figure 18.9). Mood-state-dependent effects were observable with free recall but not with recognition, which resembles the findings of Kenealy (1997). In addition, the effects on free recall were greater following the generate task than following the read task.

Other similar studies provide additional support for the notion that mood-state effects are greater when internal events (rather than external events) are involved at learning and/or test (Macaulay et al., 1993).


Most research concerned with mood effects on attention and perception has focused on anxiety or depression. One reason is that they can be studied at both normal and clinical levels. However, a problem with trying to compare the effects of anxiety and depression on cognitive functioning is that individuals who are high in anxiety tend to be high in depression, and vice versa. This is true of both normal and clinical populations.

Research in this area has been strongly influenced by Beck's schema theory, which predicts that there should be facilitated processing of schema-congruent information in attention and perception. However, Bower's (1981) network theory is also relevant. It has mainly been applied to memory, but has implications for other aspects of cognitive functioning. According to network theory, whenever the node corresponding to an emotion is activated, activation spreads out to all of the related nodes. If someone is happy, then nodes relating to happy personal experiences and similar concepts to happiness (e.g., euphoria, joy, contentment, and so on) will be activated. This widespread activation should facilitate performance across a wide range of tasks involving processing of happiness-related information.

The focus in this section will be on two main cognitive biases. First, there is attentional bias, which is selective attention to threat-related rather than neutral stimuli. Second, there is interpretive bias, which is the tendency to interpret ambiguous stimuli in a threatening rather than an innocuous fashion.

Before proceeding to discuss the effects of anxiety and depression on attention and perception, it is worth mentioning their effects on memory. The main focus of the research has been on two memory biases:

1. Explicit memory bias, in which negative or threatening information is retrieved relatively better than positive or neutral information on a test based on conscious recollection.

2. Implicit memory bias, in which memory performance for negative information is relatively better than that for neutral information on a test in which conscious recollection is not involved.

Conquering Fear In The 21th Century

Conquering Fear In The 21th Century

The Ultimate Guide To Overcoming Fear And Getting Breakthroughs. Fear is without doubt among the strongest and most influential emotional responses we have, and it may act as both a protective and destructive force depending upon the situation.

Get My Free Ebook

Post a comment