NNAT3 Level A study plan
NNAT3 Level A · Give each week a clear focus.
8 weeks · About 6 hours per week
Pattern Completion
Identify the missing region in a partly shown grid pattern by stating which rows, columns, and corners the gap touches
Describe the repeating unit of a pattern in words before choosing an answer, naming its shape, count, and orientation
Reject answer choices that mirror or rotate the repeating unit instead of continuing it
Complete 30 pattern-completion items untimed while writing a one-line reason for each choice
Your checkpoint
A 30-item error log listing each wrong answer with the rule that the correct choice followed
Pattern Completion & Reasoning by Analogy
Match paired figures in analogy items by naming the single attribute that changed between them
Classify the change as size, color, count, position, rotation, or shape substitution
Apply the same change rule to the third figure and predict the answer before looking at options
Time 25 mixed pattern-completion and analogy items at 45 seconds each
Your checkpoint
A two-column chart of 20 analogy pairs showing the attribute changed and the rule applied
Reasoning by Analogy
Distinguish attribute-change analogies from relationship analogies such as part-to-whole and inside-to-outside
Handle two-rule analogies where both shape and position shift
Eliminate options that copy the second figure instead of transforming the third
Explain aloud the rule for 15 items in under 20 seconds each
Your checkpoint
A recorded or written set of 15 spoken rule explanations, each under 20 seconds
Reasoning by Analogy & Serial Reasoning
Order a three- or four-frame sequence by identifying the step that repeats each frame
Detect sequences that alternate between two rules rather than repeating one
Predict the next frame in 20 sequences before viewing the choices
Compare analogy and serial items to state how the reasoning differs
Your checkpoint
A 20-item sequence sheet with the step rule written beside each predicted frame
Serial Reasoning
Track cumulative changes such as growing counts or rotating positions across a full sequence
Spot the frame where a sequence breaks its own rule and explain why
Solve 25 serial-reasoning items at 50 seconds each with no skipped items
Rank the four most common rule types by how often you miss them
Your checkpoint
A ranked list of your four weakest rule types with the count of misses for each
Serial Reasoning & Spatial Visualization
Mentally fold, rotate, or flip a figure and state the result before checking options
Identify which option is a rotation versus a mirror image of the same figure
Assemble a whole figure from described parts without drawing it
Complete 20 spatial-visualization items with a sketch only after answering
Your checkpoint
A 20-item log marking each answer as rotation, mirror, or assembly and noting whether you were correct
Spatial Visualization
Apply a stated fold-and-cut or fold-and-punch operation and predict the unfolded result
Track which side of a folded figure faces up after one or two folds
Solve paper-folding items by working backward from the punch to the flat sheet
Finish 25 spatial items at 55 seconds each
Your checkpoint
A 25-item accuracy score for spatial visualization with a list of every miss and its cause
Pattern Completion & Reasoning by Analogy & Serial Reasoning & Spatial Visualization
Run two full-length mixed practice sets under timed conditions
Review every miss and tag it to one of the four domains
Re-drill the two domains with the lowest accuracy from weeks 1 through 7
Confirm pacing so no item takes longer than 60 seconds
Your checkpoint
A final score report showing per-domain accuracy and a one-page list of rules to review before test day
Milestones
Week 2: Score at least 80 percent on a 25-item pattern-completion set and explain the rule for every item
Week 4: Score at least 75 percent on a 30-item mixed set covering analogy and serial reasoning
Week 6: Correctly label at least 18 of 20 spatial items as rotation, mirror, or assembly before checking
Week 8: Reach at least 80 percent overall on a timed full-length set with no domain below 70 percent