Fighting Game Test Methodology, Scoring, and Evidence

First: this is not a psychological test

The Fighting Game Test is an original model for organizing decisions that appear during fighting game matches. It is not a standardized clinical, personality, aptitude, intelligence, or reaction assessment. At present, the test has not been tested on a large representative sample for internal reliability, test-retest reliability, factor structure, or criterion validity. It would therefore be inaccurate to describe the 16 results as scientifically proven personality types.

The questions are not random. The design is informed by research topics including perception and anticipation in sport, the visual information used by experts, analytical and intuitive judgment, naturalistic decision-making under time pressure, planning, and adaptation. Those research areas do not directly validate these four axes. They provide concepts that we translated into observable fighting game situations.

The four axes

1. Act First (Extraversion) and Think First (Introversion)

This axis compares learning by entering the match and producing the next piece of information with observing and organizing an idea before acting. It is not a general measure of sociability. The fighting game questions stay close to action: testing a matchup in ranked play versus researching first, trying a correction immediately after a loss versus pausing to name the cause, and moving to expose a response versus watching before committing.

2. Concrete Detail (Sensing) and Big Picture (Intuition)

This axis compares prioritizing concrete information such as spacing, moves, frames, and inputs with prioritizing intentions, match flow, and the structure of a win condition. Both sides still use the screen. The distinction is whether a player tends to build an answer from observable details or compress several events into a larger interpretation of what the opponent is trying to do.

3. Rationality (Thinking) and Personal Meaning (Feeling)

This axis compares explainable criteria such as efficiency, expected value, and risk-reward with identity, physical feedback, and a way of winning that feels personally coherent. It does not separate logical people from emotional people. Both standards contain thought and feeling. The questions ask which standard a player returns to when several plausible choices compete.

4. Prepared Plan (Judging) and Flexible Change (Perceiving)

This axis compares reproducing a prepared win condition or response order with changing shape around the opponent and situation. Strong play requires both. The quiz asks whether a player tends to restore the planned route when it breaks or prioritize new information and update the route during the match.

The 43 questions and scoring model

Part one contains 16 questions, with four questions assigned to each axis. The five answer positions are scored from 1 to 5. The four answers on an axis are summed and normalized to a position from 0 to 100, with 50 as the midpoint. A value below 50 selects the left letter and a value above 50 selects the right letter. E/I, S/N, T/F, and J/P combine into one of 16 result labels. An exact tie opens one additional question for that axis so the result can be presented as an estimate rather than an arbitrary default. The result page describes direction and strength in words, such as “strongly favors concrete detail,” so a position near zero is not mistaken for low ability.

Part two contains 27 questions: nine for current cognitive style, five for winning aesthetic, one for the ideal player image, four for breakdown pattern, one for recovery, and seven for growth. A single-choice response contributes one vote. Growth questions rank two choices, giving three points to the first choice and one point to the second. If a category is tied and the answers do not identify one clear leader, the model falls back to the baseline associated with the four-axis type.

That separation produces two related views. The type-based baseline is inferred from the 16-question four-axis result. The measured current tendency is tallied directly from questions 17–43. A character, opponent pool, recent training focus, or temporary condition can make the two views differ. That difference is information about current play, not automatically a scoring inconsistency.

This remains a deliberately transparent additive model. It does not currently use item response theory, population norms, or probabilistic confidence intervals. A small score difference should not be interpreted as a precise ability difference. Scores near the midpoint and measured categories with close vote counts may change after a character change, a new game, or recent training.

Why the prompts describe match situations

A broad statement such as “I am a planner” invites a player to answer according to an ideal self-image. The quiz instead describes neutral, unfamiliar strings, corner pressure, resource-heavy rounds, replay review, strategy research, and resetting after losses. Concrete situations make it easier to compare an answer with actual behavior. They do not eliminate self-report bias, memory bias, social desirability, or differences in how players interpret a prompt.

Research areas that influenced the design

The following work informed the vocabulary and design questions, not the validity of the final 16-type system:

• Mann, Williams, Ward, and Janelle's meta-analysis of perceptual-cognitive expertise in sport summarizes evidence that experts often show advantages in anticipation and the extraction of visual information.
• Naturalistic decision-making and Recognition-Primed Decision research examines how experienced people recognize situation patterns and generate workable actions in time-constrained environments.
• Dual-process research discusses the relationship between fast, intuitive processing and slower, analytical processing. Modern work also identifies limitations in treating cognition as a simple binary, which is why this quiz presents continuous axis scores rather than declaring one mode universally superior.

References:

• Perceptual-cognitive expertise in sport: a meta-analysis
• Naturalistic Decision Making
• Dual-process theories of higher cognition

Limitations of a result

A result reflects self-perception at the moment of answering. It can change with the game, character, matchup, controller, online or offline environment, fatigue, recent practice, and the kinds of opponents encountered. A type label cannot determine the character a person should use, predict future rank, or establish competitive potential. Character archetypes shown on the result page are metaphors for preferred control and win conditions, not individual character recommendations.

The model also assumes that each axis can be represented through four equally weighted prompts and that the later category questions capture useful self-reported differences. Future evidence may show that some prompts are ambiguous, overlap with another construct, or behave differently across games and experience levels. The current scoring page therefore describes practical strengths and risks without claiming diagnostic precision.

Future validation and revision

After launch, any analysis of response data will occur only if anonymous collection is introduced with appropriate notice and consent. Useful checks would include item distributions, internal consistency within an axis, stability across repeated sessions, questions that produce extreme response patterns, and differences associated with game or experience level. Before data collection begins, the site will publish what is collected, why it is needed, how long it is retained, and how a user can object or request deletion where applicable.

If evidence leads to a changed question, scoring rule, or interpretation, the quiz version and update date will be documented. Until then, the best use of a result is to treat it as a question: can the tendency be observed in replays, and what changes when the player deliberately practices the other side of an axis?

Repeatable observations matter more than attractive labels, and every future revision should make that principle easier for players to test.