2 Factor Theory of Interference Methodology Retroactive Interference

2 -Factor Theory of Interference Methodology Retroactive Interference (RI): Recently learned information prevents recall of earlier learned information. EXPERIMENTAL DESIGN Experimental Group Control Group Phase 1 Phase 2 Learn List 1 Learn List 2 Learn List 1 Rest Phase 3 Test on List 1

2 -Factor Theory of Interference Methodology RI = Recall. CONTROL – Recall. EXPERIMENTAL For example, suppose there were 20 nonsense syllables in the list. The Experimental and Control Groups recall the following numbers of items: EXPERIMENTAL DESIGN Experimental Group Control Group Phase 1 Phase 2 Learn List 1 Learn List 2 Phase 3 Test on List 1 8 Learn List 1 Rest Test on List 1 17

2 -Factor Theory of Interference Methodology RI = Recall. CONTROL – Recall. EXPERIMENTAL 9 The amount of RI in the Experimental Group would be ______. Something about List 2 made the Experimental Group forget 9 items from List 1. EXPERIMENTAL DESIGN Experimental Group Control Group Phase 1 Phase 2 Learn List 1 Learn List 2 Phase 3 Test on List 1 8 Learn List 1 Rest Test on List 1 17

2 -Factor Theory of Interference Methodology Without List 2, the Experimental Group would have recalled 17 items, like the Control. Why not all 20 items? Interference theorists would say 3 items were forgotten due to uncontrolled sources of interference. EXPERIMENTAL DESIGN Experimental Group Control Group Phase 1 Phase 2 Learn List 1 Learn List 2 Phase 3 Test on List 1 8 Learn List 1 Rest Test on List 1 17

2 -Factor Theory of Interference Methodology Proactive Interference (PI): Earlier learned information prevents recall of later learned information. Suppose there were 15 items in List 2. The recall scores are as follows: EXPERIMENTAL DESIGN Experimental Group Control Group Phase 1 Phase 2 Learn List 1 Learn List 2 Phase 3 Test on List 2 6 Rest Learn List 2 Test on List 2 10

2 -Factor Theory of Interference Methodology PI = Recall. CONTROL – Recall. EXPERIMENTAL 4 PI = _____. This means that learning List 1 caused the Experimental Group to forget 4 items from List 2. EXPERIMENTAL DESIGN Experimental Group Control Group Phase 1 Phase 2 Learn List 1 Learn List 2 Phase 3 Test on List 2 6 Rest Learn List 2 Test on List 2 10

2 -Factor Theory of Interference Factor 1: Response Competition The original version of interference theory was simple. It said that interference resulted from a single process, response competition. This is when we have a tendency to make two or more responses to the same stimulus. According to the single-factor theory, when we forget something, it is because we remember something else instead. The wrong response was stronger than the right response.

2 -Factor Theory of Interference Factor 1: Response Competition For example, suppose you are given two paired-associate lists to memorize. The stimuli in the lists are identical but the responses are different: LIST 1 LIST 2 bok - xiz bok - taw Some time after you memorize List 2 you are given a test on List 1. This would be a test for _____. RI When you see bok, will you say xiz (right) or taw (wrong)?

2 -Factor Theory of Interference Factor 1: Response Competition The single-factor theory says that if you don’t say xiz, then you will say taw. LIST 1 LIST 2 bok - xiz bok - taw A response from List 2 competed with a response from List 1 and won out. Such “intrusion errors” were said to explain RI. Key prediction: The amount of RI will equal the number of intrusion errors.

2 -Factor Theory of Interference Factor 1: Response Competition For example, suppose that the Experimental Group remembered 7 items from List 1 and the Control Group 5 remembered 12. The amount of RI would be _____. LIST 1 LIST 2 bok - xiz bok - taw When we look at the responses made by the Experimental Group, we should count 5 response items from List 2. These are intrusion errors. Key prediction: The amount of RI will equal the number of intrusion errors.

2 -Factor Theory of Interference Factor 1: Response Competition Basically, theory says that all memory errors are “errors of commission”— saying the wrong thing. It does not allow for the possibility of “errors of omission” — simply saying nothing, “drawing a blank”. The prediction that RI = intrusion errors was tested in a classic experiment by Melton and Irwin (1940) with surprising results.

2 -Factor Theory of Interference Factor 1: Response Competition Classic Experiment on RI (Melton & Irwin) Phase 1: All subjects received 5 trials on a list of nonsense syllables. Phase 2: The Control Group “rested”. The Experimental Group was given a second list of nonsense syllables to memorize. The Experimental Group was divided into 4 subgroups according to how many practice trials they received on List 2: 5, 10, 20 or 40. Phase 3: All subjects got a test on List 1 about 30 minutes after they originally practiced it.

2 -Factor Theory of Interference Factor 1: Response Competition RI (Control - Experimental Classic Experiment on RI (Melton & Irwin) 4. 0 RI 3. 0 Generally, as the number of trials on List 2 increased, the amount of RI in the Experimental Group increased. 2. 0 Between 20 and 40 trials, RI slightly decreased. 1. 0 0. 0 5 10 20 Number of Trials on List 2 40

2 -Factor Theory of Interference Factor 1: Response Competition RI (Control - Experimental Classic Experiment on RI (Melton & Irwin) 4. 0 RI The number of intrusion errors fell short of the amount of RI. Between 10 and 40 trials, intrusion errors dropped sharply, falling almost to zero. 3. 0 2. 0 1. 0 0. 0 RI attributable to intrusion errors 5 10 20 Number of Trials on List 2 40

2 -Factor Theory of Interference Factor 1: Response Competition RI (Control - Experimental Classic Experiment on RI (Melton & Irwin) 4. 0 RI 3. 0 2. 0 1. 0 0. 0 RI attributable to intrusion errors 5 10 Intrusion errors decreased because as subjects got more trials on List 2, they could better discriminate between List 1 and List 2 responses. Instead of saying a List 2 response, they just didn’t say anything. 20 Number of Trials on List 2 40

2 -Factor Theory of Interference Factor 1: Response Competition RI (Control - Experimental Classic Experiment on RI (Melton & Irwin) 4. 0 RI The gap between RI and intrusion errors indicated that a second factor was present, a factor that got stronger as more trials were given on List 2. 3. 0 2. 0 1. 0 0. 0 Melton & Irwin called it Factor X. RI attributable to intrusion errors 5 10 20 Number of Trials on List 2 40

2 -Factor Theory of Interference Factor 2: Extinction Classic Experiment on RI (Melton & Irwin) Basically, as subjects tried to recall List 1 responses, they often just said nothing (errors of omission). They probably thought of List 2 responses but didn’t say them because they knew they were from List 2. According to Melton & Irwin, this may have happened because as the subjects learned List 2, they repeatedly said responses from List 1 and got no “reinforcement”– feedback that they were correct.

2 -Factor Theory of Interference Factor 2: Extinction Classic Experiment on RI (Melton & Irwin) In terms of operant conditioning, if you repeatedly make a response and get no reinforcement for it, what happens? You stop making that response. This is extinction called ______. As you get more trials on List 2, you will have more chances to make List 1 responses and be wrong. Extinction of List 1 responses should get stronger.

2 -Factor Theory of Interference Factor 2: Extinction Classic Experiment on RI (Melton & Irwin) Now you go back and try to remember those same List 1 responses. You can’t. You “draw a blank”. Melton & Irwin referred to this second factor as “unlearning”, a misleading term because extinction does not cause a response to be unlearned or erased from memory. Interference theorists maintained that these extinguished List 1 responses were still in memory, and they made a surprising prediction.

2 -Factor Theory of Interference Factor 2: Extinction Recall of List 1 responses should go UP over time! We usually think of recall as going down over time. But recall of List 1 should go UP due to “spontaneous recovery”. This usually goes along with extinction of operant responses. Unfortunately, years of research failed to demonstrate this predicted increase, and interference theory started to lose its appeal.