tex4

• Launched the modular approach of memory in 1968.
• The multi store model of memory in an explanation of how memory processes work.
• The sense organs have only a limited capacity of storing information coming from the environment through what Atkinson and Shiffrin called Sensory Register. From here, the information goes into the STM, where it is retained long enough to use it. If the information from STM is consolidated then it goes into LTM and it is stored there from minutes to a lifetime.

• Used a tachistoscope to present participants with a 3X3 matrix of letters, which they had to recall after the 50ms presentation.
• In the total recall paradigm participants were asked to report as many letters as they could. Typically only 3-4 letters were correctly recalled.
• In the partial report paradigm Sperling instructed participants to recall only a single row of the display. The signal was a tone sounded immediately after the display was turned off. A high pitch tone indicated the top row should be reported, while intermediate and low tones indicated the middle and bottom row respectively. The result was almost 100% accuracy in the reports of the critical row. However, accuracy decreased if the tone was delayed for 250ms and with delays of a second or more, it dropped to the level characteristics of ordinary full reports assessed by the total recall paradigm.
• Sperling concluded that the trace duration in the sensory register is somewhere between 250-500ms.

• Participants are given a series of chunks. Immediately after the series, participants carry out a paced and complex activity: typically, they are asked to count backwards by threes from a presented number. After a period of time, this activity is terminated and the participants are asked to recall the original series. If the distracter task is longer than 15-20s, the participants can no longer recall the stimulus material.
• The info in the STM lasts for 10-15 seconds

The interference theory states that people forget not because memories are actually lost from storage, but because other information gets in the way. Information loss in STM is due to interference.

See: Retroactive Interference and Proactive Interference

• Interference- theory states that people forget not because memories are actually lost from storage, but because other information gets in the way. They get confused.
• Retro- try to remember list A but list B gets in the way
• Pro- try to remember list B but list A gets in the way

• Refers to how items at the beginning and at the end of one list are better recalled than ones in the middle.
• They gave participants a long list of words to remember and afterwards they asked them to remember as many words as possible. Turns out that overall, people are more likely to remember items from the beginning of the list (primacy effect) and from the end of the list (recency effect).
• Primacy Effect is typically attributed to additional rehearsal of items earlier in the list (LTM). In contrast the recency effect is attributed to STM.
• Serial position effect supports the interference theory and the fact that information is lost in STM through displacement.

He used think aloud protocols and he discovered that participants used a rehearsal strategy to remember the information, especially when the primacy effect is concerned.

• He came up with the famous magic number: 7+/-2 chunks of information using the digit span task.
• Participants are presented with auditory series of digits and are asked to recall that series in the same order as in the original presentation. It turns out that people with "good" memory can recall around 9-10 digits, whereas people with "poor" memory can recall around 5 digits.
• STM has a limited capacity: 7+/-2 chunks. Also, in order to increase the memory capacity, people chunk information.

• Faloon had 35 weeks (1hr/day) of training and he increased his digit span to 80 by chunking information. When presented with a series of digits he used to chunk these digits into meaningful units.
• Yes, one can improve his digit span if improving his chunking capabilities.

• Sternberg presented participants with strings of letters (or numbers) for a couple of seconds. (e.g. QTWPBNVC)Then he asked participants to say (yes or no) whether the target letter "B" was part of the previously presented string. He measured the reaction time for each response.
• The search in STM can be either parallel or serial. The serial search can be either exhaustive or self-terminating.
• In STM the retrieval is through serial exhaustive search.

• They presented to participants 4 lists of words:
• Acoustically similar (e.g. man, map, mad...)
• Acoustically dissimilar (e.g. pen, dog, set...)
• Semantically similar (e.g. huge, big, great...)
• Semantically dissimilar (e.g. chair, soap, drink...)

• Afterwards, he asked participants to recall the words either immediately after their presentation (testing STM) or after a distracter task (testing LTM). The results show that in the STM condition, acoustically dissimilar words were better remembered than acoustically similar words, and in the LTM condition, semantically dissimilar words were better remembered than semantically similar words.
• In the STM information is represented acoustically, and in the LTM information is represented semantically.

• He asked people to learn a list of 16 words.
• After they've learned the words, he split the group in two: Group 1 was given a list of 32 words (the 16 that they studied initially + 16 new words) and Group 2 was given a list of 32 new words.
• Group 2 performed better than Group 1 because of subjective organization.