Levels of processing and Craik and Tulving (1975): WACE Year 12 Psychology Unit 3
“Levels of processing model of memory (Craik and Lockhart, 1972): shallow (structural, phonemic) and deep (semantic, elaboration) processing; study: Depth of processing and the retention of words in episodic memory (Craik and Tulving, 1975)”
The levels of processing model (Craik and Lockhart, 1972) says memory depends on how deeply information is processed, not on which store it is in. Shallow processing focuses on appearance (structural) or sound (phonemic) and produces weak, short-lasting memories. Deep processing focuses on meaning (semantic) and links it to existing knowledge (elaboration), producing more durable memories. Craik and Tulving (1975), the designated study, showed that words processed for meaning were recognised far better than words processed for appearance or sound.
Jump to a section
What this dot point is asking
You need to describe the model, classify a task as structural, phonemic or semantic, and know the aim, method, key findings, contribution and limitations of Craik and Tulving (1975).
The levels of processing model (Craik and Lockhart, 1972)
Fergus Craik and Robert Lockhart proposed an alternative to thinking about memory as a series of stores. They argued that the durability of a memory depends on the depth of processing when it is encoded.
| Level | Type of processing | Focus | Example question | Memory |
|---|---|---|---|---|
| Shallow | Structural | What the word looks like | Is it in capital letters? | Weakest |
| Shallow | Phonemic | What the word sounds like | Does it rhyme with "cat"? | Moderate |
| Deep | Semantic | What the word means | Is it a type of animal? | Strongest |
| Deep | Elaboration | Linking meaning to existing knowledge and personal experience | How does this relate to your own life? | Strongest and most lasting |
Elaboration is deep processing that connects new information to what you already know, for example by creating examples or relating it to yourself. It is the same idea behind elaborative rehearsal (see rehearsal and the forgetting curve).
Designated study: Craik and Tulving (1975)
Aim. To test whether the depth at which words are processed affects how well they are retained in episodic memory.
Method.
- Participants were shown a series of words, each very briefly, and for each word were asked a question that needed a "yes" or "no" answer.
- Structural questions: "Is the word in capital letters?"
- Phonemic questions: "Does the word rhyme with WEIGHT?"
- Semantic questions: "Would the word fit the sentence: 'He met a ___ in the street'?"
- Learning was incidental: participants did not know they would be tested.
- Afterwards they were given a surprise recognition test, picking the words they had seen from a longer list including new words.
Key findings.
- Recognition was lowest after structural processing, higher after phonemic processing, and highest after semantic processing.
- Words that received a "yes" answer were generally better remembered than words that received a "no", which suggests that words that fit the question were processed more richly.
Contribution. The study gave controlled experimental support to the levels of processing model, and showed that how information is encoded, not just how long it is rehearsed, determines how well it is remembered. It supports study techniques that focus on meaning.
Limitations.
- Circularity of "depth": depth is hard to measure independently of memory itself.
- Time and effort: semantic questions may take longer to answer, so extra processing time or effort could partly explain the result (Craik and Tulving ran further experiments to address this).
- Artificial task: single words in a laboratory, followed by a surprise test, may not reflect learning of complex, meaningful material.
- Sample: participants were university students, which limits generalisability.
Structural and phonemic processing are shallow; semantic processing and elaboration are deep. Deeper processing leads to stronger, longer-lasting memories.
Three students revising the hippocampus
- Tom highlights the word "hippocampus" in yellow every time it appears: structural (shallow) processing.
- Priya makes up a rhyme about how to say "hip-po-cam-pus": phonemic (shallow) processing.
- Jack explains in his own words how damage to the hippocampus affected Henry Molaison, and relates it to forgetting where he parked: semantic processing and elaboration (deep).
According to the model, Jack will remember the most in the exam.
Marker's note: classify each technique and justify it by what feature of the information is being processed.
- Calling phonemic processing deep
- Sound-based processing is shallow; only meaning-based processing is deep.
- Saying participants were told to memorise the words
- Learning was incidental; the recognition test was a surprise.
- Leaving out the "yes" and "no" finding
- It is part of the key findings.
- Forgetting the dates
- The model is Craik and Lockhart (1972); the designated study is Craik and Tulving (1975).
Exam-style questions
Questions in the style of SCSA exam questions on this dot point, each with a worked answer. They are written by ExamExplained unless tagged "Past paper"; the year shows the paper a question is modelled on.
Original8 marksDescribe the aim, method and findings of Craik and Tulving's (1975) study of depth of processing, and explain its contribution to psychology.Show worked answer →
Two marks each for aim, method, findings and contribution.
- Aim
- To test the levels of processing model by finding out whether words processed more deeply (for meaning) are remembered better than words processed shallowly.
- Method
- Participants were shown words one at a time, each preceded by a question requiring a yes or no answer. Structural questions asked about appearance ("Is the word in capital letters?"), phonemic questions about sound ("Does the word rhyme with WEIGHT?") and semantic questions about meaning ("Would the word fit the sentence 'He met a ___ in the street'?"). Participants were not told their memory would be tested (incidental learning). Afterwards they were given a surprise recognition test.
- Findings
- Recognition was lowest for words processed structurally, higher for phonemic processing and highest for semantic processing. Words given a "yes" answer were generally remembered better than those given a "no".
- Contribution
- It provided experimental support for Craik and Lockhart's (1972) claim that deeper, meaning-based processing produces more durable memories, and it underpins study advice to process material for meaning rather than simply repeating it.
Original4 marksExplain one limitation of the levels of processing model and one limitation of Craik and Tulving's (1975) study.Show worked answer →
Two marks for each limitation explained.
Model. "Depth" is hard to define or measure independently. It is often defined by how well something is remembered, so the explanation becomes circular: deep processing is said to cause good memory, and good memory is taken as evidence of deep processing.
Study. The task (answering questions about single words in a laboratory, followed by a surprise test) is artificial, so the findings may not generalise to everyday learning of complex material. Semantic questions may also take longer to answer, so more time or effort, rather than depth itself, could explain better memory.
Practise this
Sources & how we know this
- Psychology ATAR course Year 12 syllabus (effective from 1 January 2024) — School Curriculum and Standards Authority (SCSA)
- Psychology ATAR course: syllabus and support materials — SCSA
- Past ATAR course examinations — SCSA