WAIS-5 vs WAIS-IV: Index Scores, Subtests and Score Ranges Explained

The WAIS-IV reports four index scores and the WAIS-5 reports five, and neither Full Scale IQ (FSIQ) is the average of its indexes. The WAIS-5 (Pearson, 2024) replaces the old Perceptual Reasoning index with separate Visual Spatial and Fluid Reasoning indexes, builds its FSIQ from seven subtests instead of ten, and adds 15 ancillary indexes. This guide sets the two editions side by side using Pearson's own documents: the subtests, the score structure, what changed, what the WAIS-5 score descriptions (Very high, Above average, Average, Below average, Very low) do and do not say, and how to read the gaps between indexes. A WAIS report is a table of composite scores, percentile ranks and confidence intervals, not one number.

WAIS-IV: four index scores and a Full Scale IQ

The fourth edition, published by Pearson in 2008, covers ages 16:0 to 90:11 and is sold only to qualified professionals (Pearson lists it at qualification level C). Its core battery is 10 subtests, and every one of them counts toward the FSIQ. They roll up into four index scores:

• Verbal Comprehension (VCI): Similarities, Vocabulary, Information
• Perceptual Reasoning (PRI): Block Design, Matrix Reasoning, Visual Puzzles
• Working Memory (WMI): Digit Span, Arithmetic
• Processing Speed (PSI): Symbol Search, Coding

Because all ten core subtests feed the total, four of the ten, 40%, come from working memory and processing speed. Our look at what the WAIS-IV items measure shows how differently those ten subtests relate to general ability: Arithmetic and Vocabulary load highest on g, Coding and Symbol Search lowest.

WAIS-5: five primary indexes, and only seven subtests in the total

The WAIS-5 was published in 2024 for the same age range, with new norms collected between February 2023 and January 2024 on 2,020 people aged 16 to 90 (Winter, Trudel and Kaufman, Journal of Intelligence, 2024). The biggest structural change is that the old Perceptual Reasoning index was split in two, and working memory and processing speed each lost a subtest from the total. The five primary indexes are:

• Verbal Comprehension (VCI): Similarities, Vocabulary
• Visual Spatial (VSI): Block Design, Visual Puzzles
• Fluid Reasoning (FRI): Matrix Reasoning, Figure Weights
• Working Memory (WMI): Digit Sequencing, Running Digits
• Processing Speed (PSI): Coding, Symbol Search

The FSIQ uses seven of those ten primary subtests: Similarities, Vocabulary, Block Design, Matrix Reasoning, Figure Weights, Digit Sequencing and Coding. Visual Puzzles, Running Digits and Symbol Search are primary subtests that feed their indexes but not the FSIQ. Pearson lists about 45 minutes for the FSIQ and about 60 minutes for the ten primary subtests (both figures include Digits Forward, a required secondary subtest), against 70 minutes for the WAIS-IV core subtests and about 90 minutes for all of them. Information, Comprehension and Arithmetic have not vanished. They are still in the battery as secondary subtests, and Pearson's sample report lists 20 subtests in all, 10 primary and 10 secondary.

The arithmetic of the change is worth a line. Working memory and processing speed made up 40% of the WAIS-IV total (four of ten subtests) and make up 28.6% of the WAIS-5 total (two of seven). The same split into visual-spatial and fluid reasoning happened on the children's scale in 2014, which our explainer on WISC-V index scores covers, and both follow the broad-ability model described in our piece on CHC theory.

WAIS-IV vs WAIS-5: what changed?

Pearson's comparison flyer, brochures and product FAQ allow a plain side-by-side. The big changes are a smaller total, a split index and a larger battery.

• Published: WAIS-IV in 2008; WAIS-5 in 2024 (the Journal of Intelligence paper says August 2024). Both cover ages 16:0 to 90:11 and both sit at qualification level C
• Norm sample: 2,200 people in 13 age bands for the WAIS-IV; 2,020 people tested between February 2023 and January 2024 for the WAIS-5, 180 in each age band from 16 to 69 and 100 in each from 70 to 90
• Subtests in the battery: 15 for the WAIS-IV (10 core, 5 supplemental); 20 for the WAIS-5 (10 primary, 10 secondary), of which 15 are kept and 5 are new
• Subtests in the FSIQ: 10 for the WAIS-IV; 7 for the WAIS-5
• Index scores: 4, plus an optional General Ability Index, for the WAIS-IV; 5 primary indexes and 15 ancillary ones for the WAIS-5
• Dropped: Picture Completion and Cancellation. New: Running Digits, Set Relations, Naming Speed Quantity, Symbol Span and Spatial Addition
• Time: 70 minutes for the WAIS-IV core, about 90 for all subtests; about 45 minutes for the WAIS-5 FSIQ and about 60 for the ten primary subtests
• Delivery (WAIS-5): paper, or digital through Pearson's Q-global and Q-interactive systems

The two editions rank people alike. In a counterbalanced study of 186 people the FSIQs of the two tests correlated 0.92 (Winter, Trudel and Kaufman, Journal of Intelligence, 2024), and Pearson's summary of the technical data puts the average reliability of the WAIS-5 FSIQ at 0.97. What differs is the content of the total and the age of the norms, and the second of those is why scores differ, as the section on lower WAIS-5 scores below shows.

What are the WAIS-5 subtests, and which count toward the FSIQ?

Ten primary subtests feed the five index scores, and seven of them feed the FSIQ. Here is the map, from Pearson's overview brochure and FAQ:

• Verbal Comprehension: Similarities (in the FSIQ), Vocabulary (in the FSIQ)
• Visual Spatial: Block Design (in the FSIQ), Visual Puzzles (not in the FSIQ)
• Fluid Reasoning: Matrix Reasoning (in), Figure Weights (in; it was only a supplemental subtest on the WAIS-IV)
• Working Memory: Digit Sequencing (in; it replaces Digit Span), Running Digits (not in)
• Processing Speed: Coding (in), Symbol Search (not in)
• Secondary subtests (10): Information and Comprehension (verbal); Arithmetic and Set Relations (fluid); Digits Forward, Digits Backward, Letter-Number Sequencing, Symbol Span and Spatial Addition (working memory); Naming Speed Quantity (processing speed). Visual Spatial has no secondary subtest

Pearson describes the new subtests in short phrases. Running Digits is a new Working Memory subtest that measures auditory working memory requiring "flexibility in information processing and a progressive shift of attention"; Set Relations is a new Fluid Reasoning subtest that measures "deductive reasoning and logical thinking"; Naming Speed Quantity concerns "numerical processing fluency". Pearson's FAQ gives its reasons for the seven-subtest total: the chosen model was more strongly related to general intelligence (g) and best minimized racial and ethnic differences in the FSIQ. We found no one-line Pearson definition for Digit Sequencing, Matrix Reasoning, Figure Weights or the other kept subtests, so we do not paraphrase one.

How an index score is built

Each subtest is first converted from raw points to a scaled score (by Wechsler convention a mean of 10 and a standard deviation of 3; Pearson's sample report charts them from 1 to 19). The scaled scores in an index are added, and the sum is converted through the norm tables into a composite score on the familiar scale with a mean of 100 and a standard deviation (SD, the typical spread of scores) of 15 (the sample report's composite charts run from 40 to 160), as our guide to the 15 versus 16 SD question explains. Each composite then gets a percentile rank and a 95% confidence interval.

Pearson's WAIS-5 sample score report shows the mechanics. The examinee's Verbal Comprehension scaled scores sum to 29, which converts to a VCI of 124 (95th percentile, interval 115 to 130). The seven FSIQ subtests sum to 82, which converts to an FSIQ of 111 (77th percentile, interval 106 to 116). The five index scores in that report average 107.4, not 111. The FSIQ is a conversion of its own seven-subtest sum, not the mean of the indexes, a point our article on full-scale IQ versus index scores makes in general.

Fifteen more scores: the ancillary indexes

The WAIS-5 goes further than the WAIS-IV with 15 ancillary index scores, so a full report can carry 21 composites: five primary indexes, the FSIQ and the 15 ancillary ones. Pearson's product FAQ calls 14 of the 15 new, because the General Ability Index was already an optional WAIS-IV score. Pearson groups them in three families.

• Expanded indexes (5): wider versions of each primary index, such as Expanded Fluid, which adds Arithmetic and Set Relations to Matrix Reasoning and Figure Weights
• Domain-specific indexes (6): Verbal Reasoning, Quantitative Reasoning, Visual Working Memory, Auditory Working Memory (Registration), Auditory Working Memory (Manipulation) and Motor-Reduced Processing Speed
• Summary indexes (4): Nonverbal, Nonmotor, General Ability and Cognitive Proficiency

The General Ability Index is built from five subtests (Similarities, Vocabulary, Block Design, Matrix Reasoning and Figure Weights), leaving out working memory and speed. The Cognitive Proficiency Index is the mirror image, built from Digit Sequencing, Running Digits, Coding and Symbol Search. Pearson's flyer says the expanded indexes help when two subtests inside one primary index disagree. In the sample report the General Ability Index is 115 against an FSIQ of 111, a gap of 4 points that the report marks as not statistically significant. The Nonverbal, Nonmotor and Motor-Reduced Processing Speed indexes are aimed at examinees with expressive or motor difficulties, and Pearson's FAQ says the Cognitive Proficiency Index is the one ancillary index not used in its ability-achievement and pattern-of-strengths-and-weaknesses analyses. A substitution for a missing subtest is allowed only for the FSIQ and the Nonmotor Index, and only for one subtest from the same domain.

Reading the gaps between indexes

Index scores are meant to be compared, and the WAIS-5 report does the arithmetic. It compares each index with the mean of the five and every pair of indexes, printing a critical value (how big a gap must be to count as statistically significant at the .05 level) and a base rate (how common a gap that size is). In the sample report the critical values for the ten index pairs run from 10.60 to 14.39 points, so there is no single 15-point rule.

• Verbal Comprehension 124 vs Visual Spatial 103: gap 21, critical value 12.81, base rate 10% or less
• Verbal Comprehension 124 vs Working Memory 97: gap 27, critical value 12.47, base rate 5% or less
• Verbal Comprehension 124 vs Processing Speed 108: gap 16, critical value 14.10, significant, base rate 25% or less
• Working Memory 97 vs Processing Speed 108: gap 11, critical value 14.10, not significant

The 16-point gap is the useful one. It is statistically significant, yet Pearson prints base rates as bands (up to 5%, up to 10%, up to 15%, up to 25% or more than 25%) and this gap lands in the up-to-25% band, so it is not rare. The report also notes that base rates are reported by ability level. Significant is not the same as unusual, and neither is a diagnosis. That reading belongs to the examiner who gave the test. The children's counterpart, the WISC, is being restandardized as a sixth edition; see what Pearson has said about the WISC-6.

What do the WAIS-5 score descriptions (Very high, Above average, Average) mean?

Every WAIS-5 report carries a column headed "Qualitative Description". In the three Pearson sample reports we read, five labels appear: Very high, Above average, Average, Below average and Very low. Pearson does not print the score edges for them on its public pages, and its product FAQ says the qualitative descriptors "are only suggestions and are not evidence-based". What the samples show:

• Very high: composites of 124, 124 and 126 (US sample) and 125, 125 and 128 (Australia and New Zealand sample)
• Above average: 111 and 115 (US); 113, 111 and 117 (Australia and New Zealand); 114 and 111 (demographically referenced sample)
• Average: 97 to 109 (US); 93, 94, 95, 98 and 99 (demographically referenced sample)
• Below average: 81, 82, 84, 86 and 89 (demographically referenced sample)
• Very low: 79, printed twice (demographically referenced sample)

Those pairs fit the conventional Wechsler ladder: Very low up to 79, Below average 80 to 89, Average 90 to 109, Above average 110 to 119 and Very high from 120. That reading is our inference, not a Pearson statement: the tidy edges fit every score printed but none is stated, so check the manual before quoting a label. For a single score, see what an IQ of 110 is called on different tests.

The WAIS-IV used the older Wechsler words. The bands usually quoted are 130 and above Very Superior, 120 to 129 Superior, 110 to 119 High Average, 90 to 109 Average, 80 to 89 Low Average, 70 to 79 Borderline and 69 and below Extremely Low. Pearson's public WAIS-IV score report prints only the words Very Superior (133, 139, 145), Superior (122, 123, 128) and High Average (117) beside scores, not the table, so treat the edges as the conventional ones. For the older bands, see our IQ classification table.

WAIS-5 scores run a little lower

New norms shift scores, and this edition shows it. In a counterbalanced comparison of 186 people aged 16 to 90, the average FSIQ was 101.6 on the WAIS-IV and 99.7 on the WAIS-5, a difference of 1.9 points. The authors put the implied Flynn effect (the long-run rise in raw test performance between norm samples) at about 1.2 IQ points per decade, well below the roughly 3 points per decade usually quoted from meta-analyses. They caution that the gap may be only partly a Flynn effect, because the FSIQ itself changed from ten subtests to seven and the Working Memory index was rebuilt. A comparison of 98 sixteen-year-olds on the WISC-V and WAIS-5 (100.4 against 99.2) agreed with the adult result, but the authors warn that it does not generalize to children. The paper also reports a smaller practice gain on retest for the WAIS-5 FSIQ (3.7 points for ages 16 to 69) than the 4.5 to 5 points it cites for the WAIS-IV. Our piece on why old IQ norms inflate scores explains why a newer test tends to read lower.

What this means for online "WAIS" tests

The WAIS is an individually administered clinical instrument. It is given one-to-one by a trained examiner, takes about 45 to 60 minutes for the WAIS-5 (70 to 90 for the WAIS-IV) and is restricted to qualified professionals, so an online quiz that calls itself a WAIS test is not the Pearson instrument and cannot report its norm-based scores. What a well-built online test can do is report a score against its own stated norms, with a percentile and a range; our guides to how IQ tests work and IQ test accuracy explain what to check. The percentile calculator converts any composite on the mean-100, SD-15 scale into a rank, and our piece on the ceiling on an IQ test explains why reported scores stop near 160.

In short: read the index scores as separate measurements, treat the FSIQ as its own calculation, and read any gap against a critical value and a base rate, never a rule of thumb. The IQ Metrics test reports its score with a percentile for the same reason. For a reasoning test often compared with the WAIS, see our guide to the Watson-Glaser test; for the children's equivalent, WISC-V index scores.

Common questions