Interference occurs when two pieces of information conflict. The forgetting occurs in our LTM because we can't access our memories even though they are available.
Proactive interference is when old information interferes with the new - an older memory disrupts the newer one.
Retroactive interference is when new interferes with old - a newer memory disrupts the old one.
McGeoch and McDonald performed a study on the effects of similarity. Participants were asked to learn a list of words to 100% accuracy.
McGeoch and McDonald's study - Once the ppts could recall their first list with ease, they were given a new one with either: synonyms, antonyms, unrelated, nonsense syllables, 3-digit numbers or nothing (control)
McGeoch and McDonald found that performance depended on the nature of the second list - the most similar material produced the worst recall. Recall increased with the most different material (like the 3-digit numbers). This shows interference is strongest when memories are similar.
One strength of McGeoch et al's study on interference is that evidence from lab studies consistently demonstrate interference in memory. Most studies carried out show that both types of interference are very likely causes of forgetting from LTM. Lab experiments also control extraneous variables and so give interference a valid explanation.
Another strength of McGeoch et al's study is that it is supported by reallife studies. Baddeley et al asked rugby players to recall the no. of teams they had played in the season, week by week. Accurate reccall didn't depend on how long ago the match had taken place, but the number between them. Study shows that interference explanations can be practically applied.
One limitation of McGeoch et al's study is that they used artificial methods. The stimuli were very different from what we remember in everyday life. The use of artificial material makes interference more likely to occur in a lab, so, it may not be a cause of 'everyday' forgetting.