Proactive interference

Proactive interference occurs when old information stored in long‐term memory (LTM), interferes with the learning of new information. This usually occurs when the new information is similar to the old information. An everyday example of proactive interference is when you get a new mobile phone number: your memory for your old number will disrupt your attempts to remember your new number.

Keppel & Underwood (1962) 

Aim: To investigate the effect of proactive interference on LTM.

Method: In an experiment that is very similar to that conducted by Peterson & Peterson (1959), participants were presented with meaningless three‐letter consonant trigrams (for example, THG) at different intervals (3, 6, 9 seconds, etc.) To prevent rehearsal the participants had to count backwards in threes before recalling.

Results: Participants typically remembered the trigrams that were presented first, irrespective of the interval length.

Conclusion: The results suggest proactive interference occurred, as memory for the earlier consonants (which had transferred to LTM) interfered with the memory for new consonants, due to the similarity of the information presented.

Retroactive Interference

Retroactive interference occurs when the learning of new information interferes with the recall of old information from LTM. For example, once you have learned your new mobile number, it is often very difficult to recall your old number.

Baddeley & Hitch (1977) 

Aim: To investigate retroactive interference in everyday memory.

Method: The sample comprised rugby union players who had played every match in the season and players who had missed some games due to injury. The length of time from the start to the end of the season was the same for all players, and players were asked to recall the names of the teams they had played against earlier in the season.

Results: The players who had played the most games forgot proportionately more games than those who had played fewer games due to injury.

Conclusion: Baddeley and Hitch concluded that this was the result of retroactive inference, as the learning of new information (new team names) interfered with the memory of old information (earlier team names).

Evaluating Interference Theories

The results of Baddeley and Hitch are support by other researchers, for example McGeoch and McDonald (1931). In their experiment, participants were given a list of ten adjectives to learn (list A). Once these adjectives were learnt, participants were then given one of six other lists (list B) to learn, which varied in terms of its similarity to the original. McGeoch and McDonald found recall was worse when lists A and B were closest in similarity. This supports the idea of retroactive interference because the more similar the new material is to the previously learnt material, the greater the interference.  

Although interference research (proactive and retroactive) provides an insight into one type of forgetting, it only explains a specific type of forgetting – memory for similar information. For example, the results of Baddeley and Hitch demonstrate retroactive interference in rugby union players trying to recall team names from earlier in the season and Keppel and Underwood demonstrate proactive interference when trying to learn three‐letter consonant trigrams. Both of these examples highlight interference effects of very similar information and therefore this research is limited in its real world application and are unable to explain forgetting in other situations. 

Furthermore, interference research is often criticised for being artificial and lacking ecological validity. Most of the research examining interference is carried out in a laboratory, for example Keppel and Underwood (1962) and McGeoch and McDonald (1931), while using particularly meaningless stimuli, such as three‐letter consonant trigrams or simple word lists. As a result these findings do not represent everyday examples of interference and are limited in their application to everyday human memory.

Evaluating Context-Dependent Forgetting

Although the results suggest that environment cues aid memory recall, Godden & Baddeley didn’t control many other variables. The divers took part in the experiment at different times of the day and at different diving locations. Therefore, each diver would have experienced other contextual cues, which may have affected their memory. Therefore, we are unable to conclude whether the results of Godden & Baddeley’s research is due to the on land/underwater contextual cues, or another contextual cue provided by the different time of day or diving location.

Godden & Baddeley used a repeated measures design, as each diver took part in all four conditions. It is possible that the divers worked out the aim of the experiment and displayed demand characteristics or order effects. By the fourth trial the participants may have demonstrated practice effects where their recall improved as a result of completing the experiment multiple times, or even fatigue effects where their results declined as a result of boredom. A more suitable experimental design would have been independent measures. However, this would have required significantly more participants, which would be difficult to achieve when recruiting trained divers. Furthermore, with a sample of just 18 divers the conclusions drawn should be treated with caution. Additionally, the context examined in their study is extreme and provides little insight into context‐dependent forgetting in everyday life.

Godden & Baddeley could also be criticised for breaking ethical guidelines, in particular protection from harm. In their report, they said: ‘One diver was nearly run over during an underwater experimental session by an ex‐army, amphibious DUKW’, and therefore more precautions should have been taken to ensure the safety of their participants. 

State‐Dependent Forgetting

When the emotional state that an individual is in serves as an aid to memory recall, there is a risk that state‐dependent forgetting will occur when the same psychological state is not experienced. This is often the case with alcohol intoxication and the absence of accurate memories when sober. 

Carter & Cassaday (1998)

Aim: Carter & Cassaday (1998) examined state‐dependent forgetting using anti‐histamine drugs. These are typically administered as hay fever relief to sufferers but are also known for their sedating effect. This can make the individual feel drowsy, and therefore not as alert as they would normally be, providing a comparison to everyday non‐drug‐induced behaviour. 

Method: Participants were tasked with learning a list of words and excerpts from a text and then asked to recall the information at a later point. There were four conditions in their experiment: 

1. learn the words/text after taking anti‐histamine and recall after taking anti‐histamine

2. learn the words/text without anti‐histamine and recall without anti‐histamine

3. learn the words/text after talking anti‐histamine and recall without anti‐histamine drugs

4. learn the words/text without anti‐histamine and recall after taking the anti‐histamine drugs 

Results: In the conditions where the learning and recalling state matched (i.e. after taking the drugs on both occasions or not taking the drugs both occasions) memory was improved. Consequently, when the physiological state of the participants was different recall was significantly poorer. 

Conclusion: When the physiological/emotional cues that are present at the time of encoding are missing at the time of retrieval (recall), state‐dependent forgetting is likely to occur. 

Evaluating State‐Depending Forgetting

There is research support for the effect of state‐dependent retrieval failure, which occurs when an individual’s emotional state at the time of learning is different to their emotional state at the time of recall. For example, Goodwin et al. (1969), asked male volunteers to remember lists of words when they were either drunk or sober. The participants were then asked to recall the words 24 hours later, in either a drunk or sober state. The results of Goodwin et al. support Godden and Baddeley, as words learned when drunk were better recalled when drunk, and words learnt when sober were better recalled when sober. These results support the idea of state‐dependent retrieval failure and demonstrate the power of ‘state’ on recalling information. 

There is research support for state‐dependent forgetting using a range of different substances to create an alternative state of consciousness. Darley et al., (1973) researched the impact of marijuana on an individual’s recollection. It was found that individuals who were under the influence of marijuana when they put money in a ‘safe place’ were less able to recall where this location was once they were no longer under the influence of the drug. This evidence adds weight to the argument that the emotional and physiological state that a person is in at the time of encoding is important at the time of retrieval. 

There are issues with determining a cause and effect relationship with retrieval failure as an explanation of forgetting. Nairne (2002) criticised research in this area suggesting that there is merely a correlation between cues present at the time of encoding and cues present at the time of later retrieval. He goes further and suggests that the cues present do not in themselves cause the retrieval failure (or success), but are simply associated with it. This would mean that the cue‐dependent (context and state) explanations of forgetting due to retrieval failure, and in fact circular in nature rather than linear and psychologists are unable to conclude whether a lack of cues actually causes retrieval failure.