Memory Assessment in Children

1. Why memory assessment requires multiple tests

There are many different tests available for assessing memory in children over the age of 5. Examples include:

• Children's Memory Scale (CMS)

• Child and Adolescent Memory Profile (ChAMP)

• California Verbal Learning Test - Children's Version (CVLT-C)

• Doors and People

• Rivermead Behavioural Memory Test for Children (RBMT-C)

• NEPSY-II memory subtests

The important principle is that these tests do not all measure the same thing. Memory performance can vary according to:

• whether information is verbal/auditory or visual/non-verbal;

• whether it is tested immediately or after a delay;

• whether the child has to freely recall information or merely recognise it;

• how effectively information is initially learned/encoded;

• how information is organised during learning;

• how vulnerable memory is to interference;

• whether the task resembles everyday memory demands.

Therefore, assessment aims to establish a memory profile.

Exam principle: Do not think "What is the child's memory score?" Think "Which aspects of memory are intact and which are impaired?"

2. A useful framework for understanding memory assessment

The model shown on page 3 places memory within a wider cognitive system.

At the top is directed attention, which feeds into short-term/immediate memory. Immediate memory can involve:

• auditory/verbal information

• visual/non-verbal information

Information can then be actively maintained and manipulated through working memory.

Through learning, information can enter long-term memory.

Long-term memory can then be divided into:

Declarative memory

Memory that can be consciously accessed.

This includes:

• episodic memory - memory for personally experienced events;

• semantic memory - memory for facts and general knowledge.

Procedural memory

Memory expressed through skills and learned behaviour.

Examples include:

• skill learning

• classical conditioning

Long-term information can then be accessed through:

• free recall, where information must be independently retrieved;

• recognition recall, where the correct information is identified when alternatives or cues are provided.

This framework explains why a child can perform differently across apparently similar memory tasks. A child may have encoded information successfully but struggle to retrieve it independently, for example.

3. Encoding versus retrieval - a crucial distinction

One of the most important ideas for interpreting memory tests is distinguishing an encoding/learning problem from a retrieval problem.

Encoding difficulty

The information has not been adequately learned or stored in the first place.

If encoding is poor, providing recognition cues later may not dramatically improve performance because the memory representation itself is weak.

Retrieval difficulty

The information has been learned, but the child has difficulty independently accessing it.

A useful pattern is:

poor free recall + substantially better recognition = possible retrieval difficulty

The recognition cue helps the child access information that was already encoded.

This is why tests frequently include both free recall and recognition conditions.

For example, ChAMP explicitly includes recognition trials to help distinguish retrieval versus encoding deficits.

4. Children's Memory Scale - CMS

The Children's Memory Scale (CMS) was developed by Morris J. Cohen in 1997 and provides a comprehensive assessment of learning and memory.

Age range

5-16 years

Administration time

• Core battery: approximately 30-35 minutes

• Supplemental battery: an additional 10-15 minutes

Standardisation

The normative sample consisted of approximately 1,000 typically developing children across 10 age groups.

Around 300 children also completed the WISC-III or WPPSI-R, allowing relationships between standardised measures of intelligence and memory to be examined.

This matters because memory and general intellectual ability are related but are not interchangeable. Memory therefore needs to be assessed as its own cognitive domain.

5. CMS structure

The CMS assesses three broad domains:

1. Auditory/Verbal

Produces:

• Verbal Immediate

• Verbal Delayed

• Delayed Recognition

• Learning

2. Visual/Non-Verbal

Produces:

• Visual Immediate

• Visual Delayed

• contributes to Learning

3. Attention/Concentration

Produces an:

• Attention/Concentration Index

Information from these domains contributes to an overall General Memory index.

This design is clinically useful because a child could, for example, show:

good visual memory + weak verbal memory

rather than a general memory impairment.

6. CMS Auditory/Verbal subtests

The main auditory/verbal tasks include:

• Stories Immediate

• Stories Delayed

• Word Pairs Immediate

• Word Pairs Delayed

A supplemental Word Lists task is also available.

Stories

Two age-dependent stories are read aloud:

• versions for 5-8 years

• 9-12 years

• 13-16 years

Immediately after each story, the child repeats as much as possible.

Delayed condition

Later, the child:

1. retells the stories;

2. answers questions about them, providing a recognition component.

Why is this useful?

Story memory provides meaningful contextual information. It therefore differs from remembering unrelated individual words.

A child who remembers stories better than isolated words may benefit from semantic organisation and contextual structure.

7. CMS Word Pairs

The child hears pairs of words.

Depending on age, there are:

• 10 pairs for 5-8-year-olds

• 14 pairs for 9-16-year-olds

The pairs include both related and unrelated words.

The first word, or stem, is presented and the child must retrieve its associate.

There are three learning trials, with corrective feedback, followed by free recall.

Delayed condition

The child later:

1. spontaneously recalls the pairs;

2. completes a recognition condition.

This task therefore assesses verbal associative learning, rather than simply remembering a continuous story.

8. CMS Visual/Non-Verbal Memory

The core visual/non-verbal tasks include:

• Dot Locations Immediate

• Dot Locations Delayed

• Faces Immediate

• Faces Delayed

There is also a supplemental Family Pictures task.

Dot Locations

The child sees blue dots positioned within a rectangle.

The stimulus is removed and the child reproduces their spatial positions using chips on a grid.

The task is age-adjusted:

• 6 dots for 5-8-year-olds

• 8 dots for 9-16-year-olds

There are three learning trials.

A distractor array of red dots is then introduced, after which the child must recall the original blue-dot arrangement.

Later, in the delayed condition, the child again reproduces the original blue-dot array.

What does this assess?

This is particularly useful for examining:

• visual-spatial learning

• spatial memory

• retention

• resistance to interference

The distractor array is important because the child has to maintain the original representation despite competing visual information.

9. CMS Faces

The child sees individual photographs of human faces.

Number presented:

• 12 faces for 5-8-year-olds

• 16 faces for 9-16-year-olds

During immediate and delayed conditions, the child identifies previously presented faces among foils.

There are:

• 36 photographs for younger children

• 48 photographs for older children.

This therefore relies strongly on visual recognition memory.

10. CMS Attention/Concentration Index

Memory performance depends partly upon the child's ability to attend to and manipulate information.

Two relevant CMS tasks are:

Numbers

This is similar to WISC digit span and involves digit span forwards and backwards.

Sequences

The child must mentally sequence or manipulate information quickly.

Examples include:

• numbers;

• days of the week;

• months;

• producing sequences forwards and backwards;

• counting in 2s, 4s and 6s.

Scoring incorporates accuracy and speed.

Why does this matter?

Poor performance on a memory test does not automatically mean that long-term memory is impaired.

If the child has difficulty attending to information or holding it in working memory, the information may never be encoded effectively enough to be remembered later.

11. Child and Adolescent Memory Profile - ChAMP

The Child and Adolescent Memory Profile (ChAMP) was developed by Elisabeth M. S. Sherman and Brian L. Brooks in 2015.

It assesses both:

• visual memory

• verbal memory

and includes:

• immediate

• delayed

conditions.

Age range

5-21 years

Administration

• individually administered;

• approximately 35 minutes;

• a Screening Index can be administered in approximately 10 minutes.

Important design feature

The ChAMP uses:

• common real-life scenarios;

• colourful stimuli;

• developmentally appropriate material;

• engaging and relevant stimuli for children and adolescents.

Importantly, no motor response is required.

12. Why ChAMP is useful

Many older memory measures originated in adult neuropsychology and were subsequently adapted for children.

The ChAMP was instead deliberately designed around children and adolescents, making the material more developmentally relevant.

Its standardisation included more than 1,200 participants representative of the 2012 US population.

There was also a clinical sample of more than 200 children, including children with:

• learning disabilities;

• ADHD;

• autism spectrum disorder;

• brain injury;

• intellectual developmental disability.

The materials report strong evidence of reliability and validity. However, the normative sample is US-based, which raises an important general issue in neuropsychological assessment: clinicians should consider whether norms and test materials are appropriate for the population being assessed.

13. ChAMP structure

There are four major subtests:

Verbal

1. Lists

2. Instructions

Visual

3. Objects

4. Places

Each has immediate and delayed components.

Together, these produce several indices:

• Immediate Memory Index

• Delayed Memory Index

• Verbal Memory Index

• Visual Memory Index

• Total Memory Index

A shorter Screening Index can be produced from immediate performance on Lists and Objects.

Why is this design useful?

It allows performance to be compared in several ways.

For example:

Verbal vs visual

and

Immediate vs delayed

Therefore, a clinician can ask whether the difficulty is modality-specific, related to delayed retention, or more general.

14. ChAMP Lists

This is a standard list-learning verbal memory test.

The child is given a familiar scenario involving going for a drive and is asked to remember 16 concrete nouns representing landmarks.

The words have relatively similar levels of:

• difficulty;

• familiarity.

There are:

• Lists

• Lists Delayed

• Lists Recognition

Recognition is especially important

If the child performs poorly when independently recalling the landmarks but recognises them accurately when choices are provided, the problem may relate more strongly to retrieval.

If both recall and recognition are weak, there is greater reason to consider problems with initial encoding or storage.

15. ChAMP Instructions

This is a paragraph memory task, but it is designed to resemble an everyday situation:

a parent giving instructions to a child.

There are 24 target concrete nouns embedded within the narrative.

The child hears instructions involving everyday activities such as getting dressed, finding breakfast items, going to school and completing homework.

The score is based on whether the target items are recalled.

It includes:

• Instructions

• Instructions Delayed

• Instructions Recognition

Why compare Instructions with Lists?

This provides an important comparison between remembering:

isolated verbal information

versus

verbal information embedded within meaningful linguistic context.

Some children may improve when context provides a meaningful organisational framework.

Other children may perform worse because the additional linguistic information increases the amount of material that must be processed.

Therefore:

Better Instructions than Lists may suggest that context facilitates memory.

Worse Instructions than Lists may suggest that additional contextual/linguistic information creates excessive processing demands.

16. ChAMP Objects

The Objects task assesses visual memory for features including:

• shape;

• texture;

• visual detail;

• three-dimensional characteristics;

• man-made and natural objects.

It includes:

• Objects

• Objects Delayed

The objects are deliberately designed to be difficult to verbalise and name, while remaining colourful and visually engaging.

Why is difficulty verbalising the stimuli important?

A major problem with some supposed "visual memory" tasks is that participants can verbally label the stimuli.

For example, if someone sees a picture of an apple and thinks "apple", they could remember the word rather than relying primarily on visual memory.

The ChAMP Objects stimuli are deliberately abstract and visually similar so that the child must pay attention to actual visual characteristics.

This makes the task a cleaner assessment of visual memory.

17. ChAMP Places

The Places task assesses memory for visual scenes.

It includes:

• Places

• Places Delayed

The child must recognise:

• spatial configurations;

• contextual visual details.

The images depict everyday familiar scenes and are designed to be:

• colourful;

• realistic;

• visually interesting.

The alternatives can be extremely similar. For example, the same basic room may be presented but an object may have moved or a spatial relationship may have changed.

Why is this useful?

It assesses more than recognising the general gist of a scene.

A child may remember:

"I saw a bathroom."

but still fail to remember the precise spatial configuration of objects within that bathroom.

Therefore, the task can be sensitive to detailed visual-spatial memory.

18. California Verbal Learning Test - Children's Version - CVLT-C

The CVLT-C is especially important because it provides much more information than simply asking how many words the child remembers.

It is fundamentally an auditory verbal list-learning task.

The child hears List A, presented as a Monday shopping list.

The words are presented at approximately one word per second.

After hearing the list, the child recalls as many items as possible in any order.

This process is repeated across five immediate free-recall trials.

19. Why repeat List A five times?

Repeated trials allow the examiner to assess the child's learning curve or learning slope.

Normally, repeated exposure should produce increased recall.

For example:

Trial 1 → 6 words
Trial 2 → 8 words
Trial 3 → 10 words
Trial 4 → 12 words
Trial 5 → 14 words

The important information is therefore not just the final score.

You can examine how efficiently the child learns across repetitions.

A child whose recall remains relatively flat despite repeated exposure may have a problem with verbal learning/encoding.

20. CVLT-C List B - interference

After List A has been learned, the child hears a new List B, presented as a Tuesday shopping list.

They must learn and recall this second list.

Immediately afterwards, they are asked to return to List A.

Why introduce another list?

List B creates interference.

The assessment asks:

How well can the child maintain previously learned information after learning new, similar information?

The child must also differentiate between:

• what belonged to List A;

• what belonged to List B.

This makes the task more similar to real-world learning, where newly learned material can compete with previously learned information.

21. CVLT-C Short-Delay Free Recall

After List B, the child is asked to recall List A again without the list being re-presented.

This is the Short-Delay Free-Recall Trial.

The child must independently retrieve the Monday list while avoiding items from the Tuesday list.

Performance can be compared with List A Trial 5.

For example:

Trial 5 = 14 words
Short-delay recall = 7 words

This substantial drop may indicate vulnerability to interference or difficulty maintaining/retrieving previously learned information.

22. CVLT-C Short-Delay Cued Recall

The child is then provided with semantic category cues.

For example, they may be asked to recall:

• things to wear;

• things to play with;

• fruits.

Why is this important?

If free recall is poor but recall improves substantially when semantic cues are provided, this suggests that at least some information was available in memory but was difficult to retrieve independently.

The cues provide an organisational structure that facilitates retrieval.

Conversely, if cues provide little improvement, the problem may involve poor initial learning/encoding rather than simply inefficient retrieval.

23. CVLT-C Long-Delay Recall

Following a longer delay, List A is tested again.

The sequence includes:

Long-Delay Free Recall

The child independently recalls the Monday list.

Long-Delay Cued Recall

Semantic categories are again provided.

Long-Delay Recognition

The child hears words individually and indicates whether each word was present on the Monday list.

Together, these conditions allow the examiner to compare:

free recall → cued recall → recognition

This is extremely useful for understanding why memory performance is poor.

24. CVLT-C recognition and foils

Recognition is more sophisticated than simply mixing completely unrelated words with the target words.

Foils can have different relationships with the original material.

Examples include words that are:

• semantically related to target items;

• from the same category;

• phonemically similar, meaning they sound similar;

• previously presented on List B;

• relatively unrelated.

Why?

Different errors reveal something about the child's memory representation.

For example, incorrectly saying that apple appeared because the original list contained several fruits suggests semantic/category-based confusion.

Similarly, recognising a List B word as a List A word suggests difficulty separating competing memory traces.

25. CVLT-C errors - repetitions and intrusions

The examiner records responses verbatim because errors themselves are informative.

Repetition

A child may repeatedly say the same target:

"jacket ... grapes ... jacket ... hat ... jacket"

This suggests that they may not adequately monitor what they have already recalled.

Intrusion

An intrusion occurs when the child recalls something that was never part of the target list.

For example, if the list contained:

• bananas;

• grapes;

• watermelon;

• strawberries;

the child might incorrectly recall:

"apple"

This may occur because they remember the semantic category but incorrectly retrieve a non-target member of that category.

Intrusions therefore provide information that a simple total-correct score would miss.

26. CVLT-C semantic clustering

One of the most useful CVLT-C measures is semantic clustering.

The List A items belong to semantic categories, even though they are not necessarily presented together.

A child may spontaneously reorganise them during recall.

For example:

bananas → grapes → watermelon → strawberries

This demonstrates that the child has identified a semantic relationship and is using meaning to organise information.

Why is this a good strategy?

Memory is generally easier when information is organised into meaningful groups rather than treated as unrelated individual items.

Therefore, semantic clustering provides information about the child's learning strategy, not simply their memory capacity.

A child who does not spontaneously cluster may have relatively inefficient organisation of verbal information.

27. CVLT-C serial-order strategy

Another possible strategy is recalling words according to the order in which they were presented.

For example:

word 1 → word 2 → word 3 → word 4.

This may work, but it is generally less efficient than reorganising the information semantically.

The CVLT-C can therefore examine whether the child relies on:

• semantic organisation

• serial order

The important point is that two children could remember exactly the same number of words but be using very different learning strategies.

That difference may have important implications for intervention.

28. Primacy and recency effects

The CVLT-C can also examine where within the original list remembered words came from.

Primacy effect

Better recall for words presented at the beginning of the list.

Recency effect

Better recall for words presented at the end of the list.

For example, a child who repeatedly remembers only the final few items may show a strong recency effect.

This provides additional information about how material is being encoded and retained.

29. What does the CVLT-C actually tell you?

The major strength of the CVLT-C is that it assesses much more than auditory memory.

It provides information about:

• initial verbal learning

• learning slope across repeated trials

• free recall

• semantic cueing

• delayed retention

• recognition

• interference

• semantic clustering

• serial organisation

• primacy and recency

• intrusions

• repetitions

This is why the assessment is useful despite focusing specifically on auditory-verbal material.

It gives a detailed account of the process by which the child learns and retrieves information.

30. CVLT-C scores

Some CVLT-C measures are reported using z-scores, where:

• Mean = 0

• SD = 1

Therefore:

• 0 = exactly average

• -1 = one SD below average

• +1 = one SD above average

• -2 = two SD below average

Particularly extreme scores, such as around -2 SD, deserve attention when interpreting the overall pattern.

Importantly, some strategy scores can reflect a ratio between observed and expected behaviour. For example, semantic clustering needs to be interpreted relative to the amount of material recalled.

31. Doors and People

Doors and People was developed by Alan Baddeley, Hazel Emslie and Ian Nimmo-Smith in 1994.

Its major strength is that it directly compares:

• recall versus recognition

• visual versus verbal information

within the same test.

There are four subtests:

• Doors

• People

• Shapes

• Names

The subtests were designed to be comparable in difficulty.

It can be used as:

• a clinical tool

• a research instrument

Administration takes approximately 35-40 minutes.

The original norms are for 16+ years, although the slides report a Pearson 2006 sample covering approximately 5.1-16.1 years, N = 148.

Why is Doors and People conceptually useful?

Because modality and retrieval demand can be crossed within the same assessment.

For example, a person might show:

recognition > recall

without showing a major difference between:

visual > verbal

That would suggest the major problem is associated with retrieval demand, rather than information modality.

32. Rivermead Behavioural Memory Test for Children - RBMT-C

The Rivermead Behavioural Memory Test for Children (RBMT-C) was developed by Barbara A. Wilson, Rebecca Ivani-Chalian and Frances Aldrich in 1991.

Age range

5 years to 10 years 11 months

Administration

Approximately 25-30 minutes.

Central purpose

The RBMT-C was designed to assess and help predict everyday memory problems in children.

This makes it conceptually different from tests such as the CMS.

It was originally developed for use by occupational therapists, and therefore is not primarily driven by component theories of memory.

Instead, it asks:

Can the child's memory function effectively in situations resembling everyday life?

33. Why everyday memory matters

Traditional memory tests may ask someone to:

• learn lists;

• remember stories;

• reproduce locations.

These are useful experimentally and clinically, but everyday memory difficulties often involve things such as:

• forgetting an appointment;

• forgetting someone's name;

• losing belongings;

• forgetting where something was placed;

• forgetting to deliver a message;

• forgetting a route.

The RBMT-C was designed to bridge this gap between formal memory testing and behavioural everyday memory difficulties.

This gives the test stronger ecological relevance.

34. RBMT-C subtests

There are 12 subtests, including:

• remembering a name;

• remembering a hidden belonging;

• remembering an appointment;

• picture recognition;

• immediate prose recall;

• face recognition;

• remembering a short route;

• delivering a message;

• orientation questions.

The tasks are deliberately brief and behaviourally meaningful.

35. RBMT-C - remembering a name

The child is shown a photograph of a person and told the person's first and second name.

Later, they must remember the name.

This models an extremely common everyday demand:

meeting someone and later remembering who they are.

The delayed component makes this more representative of real-world memory than simply repeating the name immediately.

36. RBMT-C - hidden belonging

An item belonging to the child is hidden.

The child is told that when the examiner later says:

"We have finished this test"

they should ask for the belonging back.

This assesses an important form of prospective memory.

Prospective memory

Remembering to carry out an intended action in the future.

This differs from retrospective memory, which involves remembering information or events from the past.

The child must remember both:

1. that something needs to be done;

2. to perform that action when the appropriate cue occurs.

If they do not spontaneously remember, a prompt can be given asking whether there was something they were supposed to remind the examiner about.

37. RBMT-C - appointment

An alarm is set for approximately 20 minutes.

The child is told that when the alarm sounds, they should ask whether they need to see the examiner for another appointment.

Again, this tests prospective memory.

The child must retain an intention across a period in which other tasks occur.

If the child does not respond automatically when the alarm sounds, they can be prompted by asking whether there was something they were supposed to ask.

Why is this clinically relevant?

This closely resembles real-world problems such as:

• remembering an appointment;

• remembering to take medication;

• remembering to tell someone something later;

• remembering an action when a cue occurs.

38. RBMT-C - pictures

The child initially names 10 pictures.

After another task, they are shown 20 pictures and asked whether each was seen previously.

This assesses visual recognition memory.

The delay and intervening material make the task more demanding than immediate recognition.

39. RBMT-C - story/prose memory

A short story is read aloud and the child immediately repeats it.

This provides immediate prose recall.

Later, the child is asked to recall the story again, providing a delayed story recall measure.

Scoring can distinguish correctly recalled details, including partial information.

This assesses memory for meaningful verbal information rather than unrelated words.

40. RBMT-C - faces

The child is initially exposed to faces and makes simple judgements about them, such as:

• man or woman;

• old or young.

Later, additional faces are presented and the child determines whether each was seen previously.

This provides another measure of visual recognition memory.

41. RBMT-C - route memory

This is one of the clearest examples of the behavioural nature of the Rivermead.

The examiner physically demonstrates a short route around the room.

For example:

1. start at the chair;

2. walk to the door;

3. walk to the window;

4. walk to the table;

5. leave a message on the table;

6. walk back to the chair.

The child then reproduces the route.

The examiner physically acts it out rather than simply giving abstract verbal directions, although the actions may also be verbally labelled while they are performed.

Later, the child must reproduce the route again in the Route Delayed condition.

42. RBMT-C - delivering a message

The route task also incorporates remembering to pick up and leave a message.

Therefore, successful performance requires more than remembering where to walk.

The child must remember:

• the sequence of locations;

• the appropriate action;

• when and where the message must be moved.

This combines spatial/route memory with an everyday prospective action.

43. RBMT-C - orientation

The child answers 11 orientation questions.

These include information such as:

• their name;

• age;

• birthday;

• year of birth;

• day of the week;

• current month;

• current year;

• class teacher's name;

• class;

• head teacher's name;

• school.

Orientation provides a basic measure of awareness of self, time and current context.

44. RBMT-C scoring

Each subtest produces a raw score.

The raw score is converted into a standardised profile score according to the child's age.

These scores are summed to provide an:

overall profile score

The overall interpretation is relatively broad compared with more theoretically detailed memory batteries. It can indicate whether everyday memory functioning appears broadly:

• adequate;

• questionable;

• impaired.

The focus is therefore less on identifying highly specific cognitive mechanisms and more on determining whether there are meaningful everyday memory difficulties.

45. Parallel versions of the RBMT-C

The RBMT-C contains four parallel versions.

This is particularly useful for repeated assessment.

For example:

assessment before treatment → intervention → reassessment

or:

pre-surgery → surgery → post-surgery

Using exactly the same stimuli could artificially improve performance because the child remembers material from the previous assessment.

Parallel versions maintain the same type of task but change the specific stimuli.

Therefore, they help reduce practice effects.

46. RBMT-3 - adult version

The Rivermead Behavioural Memory Test - Third Edition (RBMT-3) was published in 2008.

Age range

16-89 years

It retains the everyday-memory orientation of the Rivermead but contains updated materials.

It also introduces a Novel Task assessing the ability to learn a new skill.

The individual observes a novel sequence or arrangement and must reproduce what was learned.

This extends assessment beyond remembering facts/events to include the ability to acquire a new procedure or skill.

47. NEPSY-II - Memory and Learning domain

The NEPSY-II is a broader developmental neuropsychological assessment containing multiple functional domains, one of which is Memory and Learning.

The Memory and Learning subtests shown include:

List Memory / List Memory Delayed

Ages 7-12

Assesses verbal learning and memory, including:

• rate of learning;

• interference;

• recall of verbal material.

The child hears a list of words over repeated trials and later recalls them after another list has been introduced.

This therefore shares some principles with the CVLT-C, particularly repeated verbal learning and interference.

Memory for Designs / Memory for Designs Delayed

Ages 3-16

Assesses spatial memory for novel visual material.

The child must remember both:

• the designs themselves;

• their spatial locations.

Memory for Faces / Memory for Faces Delayed

Ages 5-16

Assesses:

• encoding of facial features;

• facial discrimination;

• recognition.

Memory for Names / Memory for Names Delayed

Ages 5-16

Assesses the ability to learn the names of children over repeated trials and later remember them.

Narrative Memory

Ages 3-16

Assesses memory for organised verbal material under:

• free recall;

• cued recall;

• recognition conditions.

This allows comparison between independent retrieval and performance when additional retrieval support is provided.

Sentence Repetition

Ages 3-6

Assesses the ability to repeat sentences of increasing complexity and length.

Word List Interference

Ages 7-16

Assesses:

• verbal working memory;

• repetition;

• recall following interference.

48. Comparing the major memory assessments

This comparison is extremely useful for exam answers because the tests are complementary rather than interchangeable.

49. The most important comparisons to understand

Immediate vs delayed memory

Immediate memory asks whether information can be held/reproduced shortly after presentation.

Delayed memory asks whether the information remains accessible after time and intervening activity.

Therefore:

Good immediate + poor delayed performance may suggest difficulty retaining newly learned information over time.

Verbal vs visual memory

Verbal tasks involve material such as:

• words;

• stories;

• instructions;

• names.

Visual tasks involve:

• faces;

• spatial locations;

• visual scenes;

• abstract objects.

A discrepancy can indicate a modality-specific strength or weakness.

Recall vs recognition

Recall requires the child to generate information independently.

Recognition provides the information and requires the child to identify what was previously encountered.

Recognition generally provides more retrieval support.

Therefore:

Poor recall + relatively preserved recognition → think retrieval difficulty.

Poor recall + poor recognition → think more about inadequate encoding/storage.

This is a useful interpretative principle, but it should always be considered within the broader assessment profile rather than treated as an absolute diagnostic rule.

Learning vs retention

Repeated trials, such as the five CVLT-C List A trials, tell you about learning.

Delayed recall tells you about retention/retrieval after time.

A child may learn slowly but retain what they eventually acquire, whereas another child may learn normally but lose access to information unusually quickly.

Those are different memory profiles.

Free vs cued recall

Free recall:

"Tell me everything you remember."

Cued recall:

"Tell me all the fruits you remember."

If cues substantially improve performance, semantic structure supports retrieval.

50. Interference

Interference occurs when competing information disrupts memory.

The CVLT-C demonstrates this particularly clearly:

Learn List A repeatedly → learn List B → recall List A again.

If List A performance falls substantially after List B, the child may be particularly vulnerable to interference.

The child must also remember which information belongs to which source.

This is relevant to real learning environments because children rarely learn information in isolation. New material can compete with previously learned material.

51. Memory strategies matter

One of the biggest exam points is that neuropsychological assessment is not simply about the final score.

Consider two children:

Child A: recalls 10 words using semantic categories

Child B: recalls 10 words randomly with repetitions and intrusions

Their total recall is identical.

However, their memory processes are not identical.

Child A demonstrates an effective organisational strategy.

Child B may show inefficient organisation and monitoring.

Therefore, clinicians should consider:

• total correct;

• learning slope;

• semantic clustering;

• serial organisation;

• intrusions;

• repetitions;

• primacy/recency;

• interference;

• cueing effects;

• recognition performance.

This is why examining the process of performance is often as important as examining the numerical score.

52. Test norms and cultural relevance

Another important issue is whether normative data are appropriate for the child being assessed.

For example, ChAMP was standardised primarily using a large US normative sample.

Test materials may also contain culturally specific vocabulary. Examples in the materials include US terms such as "gas station" and "parking lot".

This matters because poor performance should reflect a memory problem, not simply:

• unfamiliar vocabulary;

• cultural differences;

• outdated stimuli;

• inappropriate norms.

Tests therefore need periodic restandardisation and updating.

This is conceptually related to why cognitive tests need contemporary normative samples as population performance changes over time.

53. Older versus newer tests

Some memory tests have been available for decades.

For example:

• RBMT-C - 1991

• CMS - 1997

• ChAMP - 2015

Older tests can remain clinically useful, but some stimuli may become visually or culturally outdated.

ChAMP represents an attempt to use more:

• colourful;

• realistic;

• age-appropriate;

• developmentally engaging;

materials.

The important issue is not simply whether a test is old, but whether its:

• norms;

• stimuli;

• cultural assumptions;

• psychometric properties

remain appropriate for the child being assessed.

54. Ecological validity versus theoretical specificity

There is an important contrast between tests such as the CMS and RBMT-C.

CMS

More strongly organised around components of memory.

It attempts to distinguish functions such as:

• verbal memory;

• visual memory;

• learning;

• delayed memory;

• attention/concentration.

RBMT-C

More strongly organised around everyday memory behaviour.

It asks whether a child can remember:

• names;

• belongings;

• appointments;

• routes;

• messages.

Therefore, the two assessments answer somewhat different questions.

CMS:

Which memory systems/processes appear relatively strong or weak?

RBMT-C:

How successfully does memory operate in everyday-like situations?

A comprehensive assessment may need both perspectives.

55. High-yield exam interpretation examples

Pattern 1

Poor free recall + good recognition

→ Information may have been encoded but is difficult to retrieve independently.

→ Think retrieval difficulty.

Pattern 2

Poor free recall + poor recognition

→ Recognition support does not rescue performance.

→ Greater concern about encoding/storage.

Pattern 3

Good immediate recall + substantially poorer delayed recall

→ Information can initially be learned but is not being effectively retained or retrieved after a delay.

Pattern 4

Poor List learning + better contextual Instructions/story memory

→ Meaningful context may facilitate organisation and retrieval.

Pattern 5

Good List learning + worse Instructions

→ Additional linguistic/contextual information may create greater processing demands.

Pattern 6

Low CVLT-C recall + strong improvement with semantic cues

→ Information is present but semantic organisation helps access it.

→ Consider retrieval/organisation difficulties.

Pattern 7

Little improvement across CVLT-C Trials 1-5

→ Weak learning slope.

→ Repetition is not producing the expected improvement in verbal learning.

Pattern 8

Good Trial 5 but substantial reduction after List B

→ Increased susceptibility to interference.

Pattern 9

Frequent List B words appearing during List A recall

→ Difficulty separating competing verbal information.

Pattern 10

Good CMS performance but poor RBMT-C performance

→ Formal memory abilities may appear relatively intact while the child has difficulty applying memory successfully to everyday demands.

This demonstrates why ecological measures can add information beyond conventional laboratory-style memory tasks.

56. Core exam takeaways

If you remember nothing else, remember this structure:

1. Memory is not one thing.
Assess verbal vs visual, immediate vs delayed, learning vs retention, and recall vs recognition.

2. Do not interpret only total scores.
Look at the child's pattern of performance and learning process.

3. Recognition helps distinguish encoding from retrieval.
Poor recall with better recognition suggests relatively greater retrieval difficulty.

4. The CVLT-C is about how verbal information is learned.
Know: 5 List A trials → List B interference → short-delay free recall → short-delay cued recall → long-delay free recall → long-delay cued recall → recognition.

5. Know the CVLT-C qualitative measures.
Especially learning slope, semantic clustering, serial recall, primacy, recency, intrusions, repetitions and interference.

6. CMS gives a broad memory profile.
Think auditory/verbal + visual/non-verbal + attention/concentration, with immediate and delayed components.

7. ChAMP is developmentally designed.
Remember the four subtests: Lists, Instructions, Objects, Places.

8. ChAMP Lists vs Instructions is an important comparison.
It asks whether context and additional linguistic information help or hinder verbal memory.

9. ChAMP Objects tries to minimise verbal mediation.
The stimuli are deliberately difficult to name so performance reflects visual memory more cleanly.

10. RBMT-C focuses on everyday memory.
Think name, belonging, appointment, pictures, story, faces, route, message, orientation.

11. RBMT-C includes prospective memory.
Remembering to ask for a belonging or respond to an alarm means remembering to perform a future action.

12. No single test gives the whole answer.
The aim is to integrate different measures to identify a child's specific pattern of memory strengths and weaknesses and understand how that relates to their functioning in everyday life.

Children’s Memory Scale (CMS)

1. The theoretical basis of memory assessment

Memory can be divided into several related but separable processes.

The model shown on page 1 begins with directed attention. Information first needs to be attended to before it can enter short-term/immediate memory. Short-term memory can involve:

• Auditory/Verbal information

• Visual/Nonverbal information

Information can then be actively held or manipulated through working memory.

Through learning, information can become part of long-term memory.

Long-term memory can be separated into:

• Declarative memory

• Procedural memory

Declarative memory

Declarative memory refers to information that can be consciously accessed and reported.

It includes:

• episodic memory - memory for events;

• semantic memory - memory for facts.

Procedural memory

Procedural memory refers more to learning expressed through behaviour or skill rather than conscious recollection.

It includes:

• skill learning

• classical conditioning

Information can later be accessed through retrieval, which can occur through:

• free recall

• recognition recall

This model is relevant because neuropsychological assessment attempts to pull these components apart. A child may have difficulty remembering verbal material but not visual material, may recall poorly but recognise accurately, or may have difficulty learning information despite intact attention. The CMS was designed around these types of dissociations.

2. Children’s Memory Scale - overview

The Children’s Memory Scale was developed by Morris J. Cohen in 1997.

It provides a comprehensive assessment of:

• learning

• memory

Age range

5 to 16 years

Administration time

• Core subtest battery: approximately 30-35 minutes

• Supplemental battery: approximately a further 10-15 minutes

Normative sample

The CMS was standardised on approximately 1,000 typically developing children across 10 age groups.

Around 300 children also completed either the WISC-III or WPPSI-R.

This additional testing allows memory scores to be compared with general intellectual ability. In particular, the CMS allows an Ability-Memory Discrepancy Analysis, where observed memory performance can be compared with the level of memory performance expected from the child’s broader cognitive ability.

Why this is relevant

The important principle is that memory ability and general intellectual ability are related but are not identical.

A child may have an average overall IQ but a significant weakness in a particular aspect of memory, or may have relatively strong memory compared with what would be predicted from their overall ability.

3. Overall CMS structure

The CMS contains three main domains:

1. Auditory/Verbal

2. Visual/Nonverbal

3. Attention/Concentration

These contribute to several index scores.

Auditory/Verbal domain

This contributes to:

• Verbal Immediate

• Verbal Delayed

• Delayed Recognition

• Learning

Visual/Nonverbal domain

This contributes to:

• Visual Immediate

• Visual Delayed

• Learning

Attention/Concentration domain

This contributes to the:

• Attention/Concentration Index

Several of these indices then contribute to an overall:

• General Memory Index

Why this matters

The CMS therefore gives both:

• specific information about different aspects of memory, and

• an overall measure of general memory functioning.

The overall score can be useful, but the profile across the different indices is often more informative clinically.

For example:

Average General Memory does not necessarily mean that every aspect of memory is average.

A child could have strong visual memory and weaker verbal delayed recognition, with the strengths and weaknesses partly averaging out.

4. CMS core and supplemental subtests

The CMS contains core subtests and supplemental subtests.

Auditory/Verbal subtests

Core

• Stories Immediate

• Stories Delayed

• Word Pairs Immediate

• Word Pairs Delayed

Supplemental

• Word Lists

Visual/Nonverbal subtests

Core

• Dot Locations Immediate

• Dot Locations Delayed

• Faces Immediate

• Faces Delayed

Supplemental

• Family Pictures

Attention/Concentration subtests

Core

• Numbers

• Sequences

Supplemental

• Picture Locations

5. What each CMS subtest measures

Stories Immediate

Measures:

• short-term memory for auditory/verbal material;

• immediate recall of meaningful verbal information.

Stories Delayed

Measures:

• consolidation

• storage

• retrieval

of newly learned verbal material.

It also includes a recognition component.

Word Pairs Immediate

Measures:

• short-term/working memory for auditory material;

• auditory learning ability;

• learning of associations between words.

Word Pairs Delayed

Measures:

• consolidation;

• storage;

• retrieval of learned verbal associations;

• recognition.

Word Lists

Measures:

• verbal learning;

• delayed auditory/verbal memory;

• delayed recognition.

Dot Locations Immediate

Measures:

• short-term/working memory for visual material;

• visual learning ability.

Dot Locations Delayed

Measures:

• consolidation;

• storage;

• retrieval of newly learned visual information.

Faces Immediate

Measures:

• immediate visual memory.

Faces Delayed

Measures:

• consolidation;

• storage;

• retrieval of visual information.

Family Pictures

Measures:

• verbal and visual experiential/incidental learning and recall.

Numbers

Measures:

• processing speed

• working memory

Sequences

Measures:

• processing speed;

• working memory;

• mental sequencing/manipulation.

Picture Locations

Measures:

• visual-spatial processing;

• immediate visual/nonverbal memory for spatial location.

6. Verbal Memory Indexes

The CMS provides:

• Verbal Immediate

• Verbal Delayed

• Delayed Recognition

Two important verbal tasks are:

• Stories

• Word Pairs

These assess different forms of verbal memory.

Stories provide meaningful contextual material, whereas Word Pairs assess associative learning.

This distinction is useful because some children may benefit from narrative context while others may perform better with more discrete information.

7. Stories - Immediate Recall

Two stories are read to the child.

The stories are age dependent, with separate material for:

• 5-8-year-olds

• 9-12-year-olds

• 13-16-year-olds

The older the child, the more complex the material becomes.

Immediately after each story is presented, the child is asked to repeat as much of the story as possible.

What is being assessed?

The task assesses:

• immediate auditory/verbal memory;

• ability to encode meaningful verbal information;

• retention of both the gist and specific details.

8. How Stories Immediate is scored

The example on page 6 shows that the examiner records the child’s response and compares it with specific scoring criteria.

The scoring does not simply depend on whether the child reproduces the story word for word.

Credit can be given for:

• specific details;

• acceptable variants;

• evidence that the child has captured the correct idea.

For example, a child may not reproduce the exact wording but may still convey that:

• the balloon stayed in the air;

• a friend accompanied the pilot;

• they travelled a particular distance;

• the balloon eventually landed.

The response therefore needs to be recorded carefully enough to determine whether the child reproduced the required information or an acceptable variant.

Why this is relevant

A child can retain the general meaning of a story while forgetting individual details.

That distinction may be clinically useful because it indicates that memory is not simply "all or nothing".

9. Stories - Delayed Recall and recognition

Following a delay, the child is asked to retell the stories again.

After delayed recall, the child answers questions about the story.

These questions provide a recognition component.

Why compare delayed free recall with recognition?

This helps distinguish whether poor delayed performance reflects difficulty:

• storing the information;

• retaining it;

• or independently retrieving it.

If the child cannot freely recall a detail but can identify it correctly when questioned, the information may still be stored but difficult to retrieve without support.

10. Word Pairs

Word Pairs assesses auditory associative learning.

A list of paired words is presented.

Depending on the child’s age, there are:

• 10 pairs for 5-8-year-olds

• 14 pairs for 9-16-year-olds

The pairs contain both:

• related word pairs

• unrelated word pairs

Examples of naturally related associations may be easier because the words already have a meaningful relationship.

Unrelated pairings are more difficult because the child must form a new arbitrary association.

11. Word Pairs procedure

The examiner first reads the word pairs.

Later, the examiner gives the first word - the stem - and the child must provide the word that originally went with it.

For example:

examiner gives the first word → child recalls the associate.

If the child gives an incorrect response or does not know the answer, corrective feedback is provided.

This means the child is told the correct associate.

There are:

• three learning trials

• followed by free recall

The child is then told to keep remembering the pairs because they will be tested again later.

12. Why corrective feedback matters

Corrective feedback allows the child to continue learning across repeated trials.

The examiner can then examine the child’s learning trajectory.

For example:

Trial 1 → 4 correct
Trial 2 → 8 correct
Trial 3 → 11 correct

would indicate improving learning across repetition.

In contrast:

Trial 1 → 5 correct
Trial 2 → 5 correct
Trial 3 → 6 correct

would show a much flatter learning curve.

Why this is relevant

The question is not only:

"How many pairs did the child remember?"

but also:

"Does the child benefit from repetition and feedback?"

A child who does not improve much despite repeated exposure may have difficulty with verbal learning or encoding.

13. Related versus unrelated Word Pairs

Some pairs have a natural or obvious relationship, whereas others are less connected.

Related pairs should generally be easier because semantic association provides a retrieval cue.

Unrelated pairs require the child to form a new association.

This means performance can give some indication of how much the child benefits from existing semantic organisation.

14. Word Pairs - delayed memory

Following a delay, the child is asked to recall the word pairs spontaneously.

This is followed by a recognition section.

This allows comparison between:

• learning across repeated trials;

• spontaneous delayed recall;

• recognition.

Again, a pattern of poor spontaneous recall with stronger recognition may suggest that the information was learned but is difficult to access without cues.

15. Visual/Non-Verbal Memory Indexes

The main visual/non-verbal core tasks are:

• Dot Locations

• Faces

Each contains:

• an immediate component;

• a delayed component.

These tasks help determine whether difficulties seen on verbal tasks reflect a general memory problem or are specific to auditory/verbal information.

16. Dot Locations - basic task

The child is shown a pattern of blue dots within a rectangle in a stimulus book.

The display is then removed.

The child must reproduce the spatial arrangement by placing chips onto a grid.

The grid differs according to age:

• 3 × 4 or 4 × 4 rectangular grid

• 6 dots for 5-8-year-olds

• 8 dots for 9-16-year-olds

This is primarily a test of visual-spatial learning and memory.

17. Dot Locations - learning trials

The original blue-dot array is presented over three learning trials.

After each presentation, the child attempts to reproduce the locations using chips.

Why three trials?

The repeated presentations allow assessment of visual learning over time.

The same principle applies as in Word Pairs:

Does performance improve across repeated exposure?

A child may not reproduce the pattern accurately on the first attempt but may gradually learn it.

18. Dot Locations - distractor array

After the three learning trials, the child is shown a different arrangement of red dots.

This is a distractor.

The child then has to reproduce the original blue-dot configuration.

Why is the distractor included?

The distractor introduces interference.

The child must:

1. remember the original spatial pattern;

2. resist confusing it with the new red-dot arrangement.

This makes the task more informative than simply assessing immediate copying.

A child who learned the original layout but becomes confused following the distractor may have difficulty maintaining information in the presence of interference.

19. Dot Locations - delayed trial

Following a delay of approximately 20 minutes, the child is asked to reproduce the original blue-dot array again.

This assesses whether the spatial information has been:

• consolidated;

• stored;

• successfully retrieved after time.

Important interpretative comparison

Compare:

performance at the end of learning

with

delayed performance

If the child performs well after repeated learning but poorly after the delay, this may indicate a problem with delayed retention or retrieval rather than initial learning.

20. Strategy during Dot Locations

The child can place the chips in any order.

The scoring focuses on whether the final positions are correct, rather than the order in which the child places them.

However, observing the strategy can still be informative.

A child may:

• use a clear systematic spatial strategy;

• mentally visualise the pattern;

• place chips randomly and repeatedly check them;

• use a different placement order each time.

Although placement order is not necessarily formally scored, behaviour during the task can provide useful qualitative information about how the child approaches visual learning.

21. Faces

The Faces subtest assesses visual recognition memory.

The child is presented with human faces one at a time.

The number depends on age:

• 12 faces for 5-8-year-olds

• 16 faces for 9-16-year-olds

The child is told to look carefully at each face and remember what it looks like.

22. Faces - immediate recognition

After the target faces are presented, the child sees additional faces.

For each face, the child indicates whether it is:

• one they saw before;

• or not one they saw before.

The target faces must be distinguished from foils.

The full recognition sets contain:

• 36 coloured photographs for younger children;

• 48 coloured photographs for older children.

This assesses visual familiarity and recognition.

23. Faces - delayed recognition

The recognition task is repeated after a delay.

The child must again identify the previously presented faces.

Comparing immediate and delayed recognition provides information about how well unfamiliar visual information is retained over time.

Why Faces is useful

Face recognition is a relatively distinct type of visual memory.

A child may perform normally on spatial memory tasks such as Dot Locations but have difficulty recognising previously presented faces, showing that even within "visual memory" there can be important dissociations.

24. Attention/Concentration Index

The CMS also assesses Attention/Concentration because poor attention or working memory can interfere with memory encoding.

The main core tasks are:

• Numbers

• Sequences

This is important because apparent memory problems can sometimes arise because the child was unable to attend to or maintain the information effectively in the first place.

25. Numbers

Numbers resembles Digit Span from the WISC.

It contains:

• digits forwards;

• digits backwards.

Digits forwards requires the child to maintain auditory information.

Digits backwards places greater demand on working memory, because the information must be held and mentally manipulated.

The task therefore assesses aspects of:

• attention;

• immediate auditory memory;

• working memory.

26. Sequences

Sequences requires the child to mentally sequence or manipulate information as quickly as possible.

The 12 items include tasks such as:

• reciting numbers;

• saying the days of the week;

• saying months of the year;

• producing them forwards;

• producing them in reverse order;

• counting by 2s

• counting by 4s

• counting by 6s

Scoring is based on:

• accuracy

• speed

For example, being asked to say the months from December backwards to January requires both automatic knowledge and active sequencing/manipulation.

27. Why Attention/Concentration is part of a memory assessment

Successful memory requires adequate initial encoding.

Encoding itself depends upon:

• attention;

• working memory;

• concentration.

If a child cannot hold information long enough to process it, later poor recall does not necessarily reflect a primary long-term memory problem.

Therefore, attention/concentration scores help clinicians interpret poor memory performance more carefully.

28. CMS scoring - scaled scores

Individual core subtests are converted from raw scores to scaled scores.

Scaled scores are centred around:

• Mean = 10

Therefore:

• around 10 = average;

• scores above 10 = relatively stronger;

• scores below 10 = relatively weaker.

The score sheet on page 11 shows an example profile in which most scores are around or above the expected range, but one lower score stands out.

The key example is a Delayed Recognition scaled score of 7.

This is approximately three scaled-score points below the mean of 10, making it a relative weakness compared with the child’s other performances.

29. Why a score of 7 should not automatically be called impaired

A score can be relatively weak compared with the child’s other scores without being severely impaired in absolute terms.

This is why interpretation considers:

• normative standing;

• percentile;

• the child’s overall profile;

• other test results;

• whether the difference is statistically significant.

The relevant question is often:

Is this a true clinical weakness, or simply the lower end of an otherwise normal profile?

30. Example CMS profile

The example child was approximately 14 years 8 months old.

The profile contains the following index scores:

• Visual Immediate = 112

• Visual Delayed = 97

• Verbal Immediate = 97

• Verbal Delayed = 94

• General Memory = 101

• Attention/Concentration = 125

• Learning = 106

• Delayed Recognition = 94

Percentile ranks shown include approximately:

• Visual Immediate = 79th percentile

• Visual Delayed = 42nd

• Verbal Immediate = 42nd

• Verbal Delayed = 34th

• General Memory = 53rd

• Attention/Concentration = 95th

• Learning = 66th

• Delayed Recognition = 34th

31. Interpreting the example profile

The most obvious strength is:

Attention/Concentration = 125

This corresponds approximately to the 95th percentile.

This is therefore a clear relative strength.

General Memory is around 101, which is close to the normative mean.

Verbal Delayed and Delayed Recognition are lower, at around 94, but still broadly within the expected range.

Important principle

The overall profile suggests:

• broadly average general memory;

• relatively strong attention/concentration;

• stronger immediate visual memory;

• comparatively weaker delayed verbal/recognition performance.

The strength of a neuropsychological profile comes from examining relative patterns, not simply deciding whether each individual score is "normal" or "abnormal".

32. Subtest-level interpretation matters

A low index score may be driven by a specific subtest.

For example, in the sample profile the lower Delayed Recognition performance appears particularly related to Stories delayed recognition, where the scaled score is 7.

However, recognition of Word Pairs is considerably stronger.

Therefore, it would be inaccurate to conclude that the child has a global recognition memory problem.

Instead, the weakness appears more specific to recognition of narrative/story information.

This illustrates why clinicians should look beneath the composite index and examine the individual subtests contributing to it.

33. General Memory Index

Several component scores combine to produce the General Memory Index.

The example gives a General Memory score of approximately:

101

This is around the normative mean.

Why the General Memory score can be misleading if used alone

A score of 101 could tempt someone to conclude:

"Memory is completely average."

However, the profile also contains:

• Visual Immediate = 112

• Attention/Concentration = 125

• Verbal Delayed = 94

• Delayed Recognition = 94

These differences tell us much more about how the child processes information.

Therefore:

Always interpret General Memory alongside the component indices.

34. Ability-Memory Discrepancy Analysis

A particularly important CMS feature is the Ability-Memory Discrepancy Analysis.

The analysis compares a child’s broader intellectual ability with their actual memory performance.

In the example, the child’s WISC-IV FSIQ = 97.

This ability score can be used to predict what level of memory performance would generally be expected.

The table then compares:

• Ability Standard Score

• Predicted Memory Score

• Actual Memory Score

• Ability-Memory Difference

• Statistical significance

• Frequency of the difference in the normal sample

35. Predicted versus actual memory

For example, if a child has:

FSIQ = 97

and based on this ability level the predicted memory score is:

99

but the actual memory score is:

112

then the child has performed better than predicted.

Conversely, if actual memory were substantially below the predicted value, this could indicate a relative memory weakness compared with the child’s general cognitive ability.

36. Example Ability-Memory Discrepancy profile

The example shown on page 13 includes approximately:

The largest difference is in Attention/Concentration.

The child’s score of 125 is much higher than the approximately 98 predicted from FSIQ.

37. Statistical significance of discrepancy scores

A large-looking numerical difference does not automatically mean that the discrepancy is clinically meaningful.

The discrepancy analysis therefore also reports whether the difference reaches a chosen statistical significance level, such as:

p < .05

Most of the example differences are marked NS, meaning not significant.

The Attention/Concentration discrepancy is sufficiently large to stand out.

Why this matters

You should not interpret every difference between IQ and memory as evidence of impairment or strength.

Differences can occur normally due to:

• measurement error;

• ordinary individual variability;

• different cognitive strengths and weaknesses.

Therefore, significance testing is necessary.

38. Base rates - how common is the discrepancy?

Even if a difference is statistically significant, it is also important to ask:

How frequently does this difference occur in the normal population?

The example indicates that the strong Attention/Concentration discrepancy occurs in approximately 10-15% of the normal comparison group.

Why base rates matter

A statistically significant difference is not necessarily rare.

If a particular discrepancy occurs in 10-15% of healthy children, it may be notable but is not extremely unusual.

This is a key neuropsychological principle:

Statistical significance and clinical rarity are not the same thing.

39. Interpretation of strengths and weaknesses

CMS interpretation should consider both:

Normative weakness

Is the child performing below age expectations?

and

Relative weakness

Is the child performing significantly below their own level of ability or below their other memory skills?

These are different.

For example:

• scaled score 7 may be lower than the child’s other scores;

• but it may not indicate a severe normative impairment.

Similarly, a score of 125 may be a clear normative strength and may also be substantially higher than expected from IQ.

40. Immediate versus delayed memory

One of the key CMS comparisons is:

Immediate → Delayed

Immediate performance tells us how well the child initially encoded and recalled information.

Delayed performance tells us how much information can be accessed after time has passed.

Example patterns

Poor immediate + poor delayed

May indicate difficulty learning/encoding the information initially.

Good immediate + poor delayed

Raises more concern about retention, consolidation or later retrieval.

Poor immediate + relatively good delayed

May suggest variable attention or initial performance rather than rapid forgetting.

These patterns should always be interpreted alongside recognition and learning scores.

41. Recall versus recognition

Another central CMS distinction is between:

• free recall

• recognition

Recall is more demanding because the child must independently generate the information.

Recognition provides additional support.

Therefore:

Poor free recall + better recognition

May indicate a greater problem with retrieval.

Poor recall + poor recognition

Raises greater concern that the information was not effectively encoded or stored.

However, this is an interpretative pattern rather than an absolute diagnostic rule.

42. Verbal versus visual memory

The CMS deliberately separates:

• Auditory/Verbal

• Visual/Nonverbal

This is important because children may have modality-specific strengths or weaknesses.

For example:

Strong visual + weak verbal

May indicate that visual learning strategies could support the child.

Strong verbal + weak visual-spatial

May indicate the child benefits more from verbal mediation than spatial presentation.

A modality comparison can therefore inform both diagnosis and recommendations.

43. Learning versus delayed memory

Learning scores and delayed scores answer different questions.

Learning

Asks:

Does performance improve through repeated exposure and corrective feedback?

Delayed memory

Asks:

Once information has been learned, how well is it retained and retrieved later?

A child can therefore show:

• slow learning but good eventual retention;

• normal learning but poor delayed memory;

• weaknesses in both.

These patterns have different clinical implications.

44. Why repeated trials are important

Repeated trials appear in both:

• Word Pairs

• Dot Locations

They allow the examiner to observe whether performance:

• improves steadily;

• improves rapidly and then plateaus;

• remains flat;

• becomes inconsistent.

This provides information about the learning process itself, not simply final memory capacity.

45. Why qualitative observation remains important

Formal scoring is essential, but behaviour during assessment can add useful information.

Examples include whether the child:

• uses a systematic strategy;

• appears to visualise information;

• organises responses;

• becomes confused following interference;

• benefits noticeably from corrective feedback;

• recognises their own errors;

• uses an inefficient random approach.

For example, during Dot Locations a child might reproduce the correct arrangement but place the chips in a different and apparently random order each time.

The final score may still be correct, but the observed strategy provides additional information about how the child is learning.

46. Important limitation - age of the CMS

The CMS was published in 1997.

Its comparison measures were based on older versions such as:

• WISC-III

• WPPSI-R

This is important when considering contemporary assessment because normative standards and comparison measures may become outdated over time.

The CMS can still provide useful information, but the age of the test and its normative framework should be kept in mind when interpreting results.

47. High-yield exam comparison of the CMS indices

Verbal Immediate

How well can the child immediately remember auditory/verbal information?

Main tasks:

• Stories Immediate

• Word Pairs Immediate

Verbal Delayed

How well can the child retain and retrieve verbal information after a delay?

Main tasks:

• Stories Delayed

• Word Pairs Delayed

Delayed Recognition

Can the child identify previously learned verbal information when retrieval support is provided?

Useful for examining:

• recall versus recognition;

• possible retrieval difficulties.

Visual Immediate

How well can the child immediately remember visual/nonverbal information?

Main tasks:

• Dot Locations

• Faces

Visual Delayed

How well can the child retain visual information across a delay?

Learning

How effectively does the child acquire information across repeated exposure?

Relevant tasks include:

• Word Pairs

• Dot Locations

Attention/Concentration

Can the child attend to, maintain and manipulate information?

Main tasks:

• Numbers

• Sequences

General Memory

Composite measure of overall memory performance.

Should always be interpreted alongside the more specific index profile.

48. Example exam interpretation

Imagine a child shows:

• General Memory = 100

• Visual Immediate = 115

• Verbal Immediate = 95

• Verbal Delayed = 90

• Delayed Recognition = 110

A good interpretation would be:

The child has broadly average general memory but shows a relative strength in visual immediate memory and weaker verbal recall. The improvement from poorer verbal delayed recall to stronger recognition suggests that at least some verbal information may have been encoded but is difficult to retrieve independently. The profile therefore suggests a possible verbal retrieval weakness rather than a global memory impairment.

This is much better than simply saying:

"Their memory is average."

49. Another exam example - learning problem

Suppose:

• Word Pairs Trial 1 = weak

• Trial 2 = weak

• Trial 3 = weak

• Delayed recall = weak

• Recognition = weak

This pattern suggests that the child is failing to benefit sufficiently from repetition and corrective feedback.

The key difficulty is therefore likely to involve initial verbal learning/encoding, rather than simply delayed retrieval.

50. Another example - retention problem

Suppose:

• Dot Locations improves strongly across the three learning trials;

• immediate post-distractor performance remains good;

• delayed recall is substantially lower.

This indicates that the child can learn the visual-spatial pattern, but access to it declines after the delay.

The problem is therefore more consistent with delayed retention/retrieval than with initial visual learning.

51. Another example - interference

Suppose the child learns the blue-dot array well but performs poorly after viewing the red-dot distractor.

This suggests that competing visual information is disrupting the original representation.

The interpretation should therefore consider susceptibility to interference.

52. Core strengths of the CMS

The CMS is valuable because it is relatively theory driven.

It attempts to separate:

• verbal from visual memory;

• immediate from delayed memory;

• learning from retention;

• recall from recognition;

• memory from attention/concentration.

This means it can provide a detailed memory profile rather than a single undifferentiated memory score.

53. Key limitations

Important limitations include:

• the test was published in 1997;

• some normative comparisons rely on older Wechsler scales;

• the full battery is relatively lengthy;

• performance on memory tasks can still be influenced by attention, language, motivation and strategy use;

• composite scores may obscure meaningful subtest-level variation.

Therefore, CMS results should be interpreted in the context of the wider neuropsychological assessment.

54. Essential exam points to remember

• CMS = Children’s Memory Scale, Cohen, 1997.

• Age range = 5-16 years.

• Core battery = approximately 30-35 minutes.

• It is a theory-driven comprehensive assessment of learning and memory.

• Main domains are Auditory/Verbal, Visual/Nonverbal, and Attention/Concentration.

• Main indices include Verbal Immediate, Verbal Delayed, Delayed Recognition, Visual Immediate, Visual Delayed, Learning, Attention/Concentration, and General Memory.

• Stories assess memory for meaningful verbal narratives.

• Word Pairs assess verbal associative learning using repeated trials and corrective feedback.

• Dot Locations assesses visual-spatial learning, interference and delayed retention.

• Faces assesses visual recognition memory.

• Numbers and Sequences assess attention, processing speed and working memory.

• Immediate vs delayed tells you about learning versus later retention.

• Recall vs recognition helps distinguish retrieval difficulty from weaker encoding/storage.

• Repeated trials allow examination of the learning curve.

• Scores should be interpreted as a profile, not individually.

• General Memory can appear average even when important strengths and weaknesses are present.

• The Ability-Memory Discrepancy Analysis compares actual memory performance with performance predicted from general cognitive ability.

• A statistically significant discrepancy should also be interpreted using its base rate, because some discrepancies occur commonly in healthy children.

• The most important clinical question is not simply "Is memory impaired?", but "Which component of memory is relatively strong or weak, and why?"