Base Color
Enter any 6-digit hex to analyse its psychology.
Context
Context affects how the brand-fit score is calculated.
Sensitivity & Environment
User sensitivity to bright colors 0.6
Environment brightness 0.5
Same HEX, different humans — these sliders tweak interpretation.
Advanced Settings
Quick Actions
Color Psychology Analysis

Color psychology is probabilistic, not destiny. Context, typography, motion, copy and culture all modulate how a color is actually felt.

Actions
Copy & Export
JSON includes all perceptual metrics, emotion scores, brand-fit data, and cultural notes.
Batch Analysis

Enter one HEX per line (max 20). Computes emotion score, warmth index, and brand-fit for each.

Click Run Batch to analyse colors.
Color Psychology Standards
Arousal-Valence Model (Russell, 1980)

Circumplex Model of Affect — Russell's two-dimensional model places emotional states on two axes: valence (pleasant–unpleasant) and arousal (activated–deactivated). Colors are assigned coordinates in this 2D space based on hue and lightness. High-chroma warm hues (reds, oranges) score high arousal; cool low-chroma hues (blues, grays) score low arousal. Valence tracks lightness more than hue — lighter tints tend positive, dark shades neutral-to-negative.

Reference: Russell, J.A. (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39(6), 1161–1178.

Valdez-Mehrabian Color-Emotion Study (1994)

Systematic HSL mapping to emotion — Valdez & Mehrabian measured pleasure, arousal and dominance (PAD) responses to 76 colors varying in hue, saturation and lightness. Key findings: saturation drives arousal more than hue; lightness drives pleasure most strongly; hue effects are secondary once saturation and lightness are controlled. High-saturation, high-lightness colors maximize pleasure+arousal simultaneously.

Reference: Valdez, A. & Mehrabian, A. (1994). Effects of color on emotions. Journal of Experimental Psychology: General, 123(4), 394–409.

Ou et al. Color Emotion Model (2004)

Cross-cultural emotion scaling — Ou et al. surveyed participants in the UK and China using 12 color-emotion scales (warm-cool, heavy-light, active-passive, etc.) applied to 218 Munsell color samples. The study produced quantitative regression equations mapping CIELAB L*, C*ab, hab coordinates to each emotion scale value. Warmth correlates strongly with long-wavelength hues (red, orange, yellow) and is modulated by chroma; weight correlates inversely with lightness.

Reference: Ou, L.C., Luo, M.R., Woodcock, A. & Wright, A. (2004). A study of colour emotion and colour preference. Part I: Colour emotions for single colours. Color Research & Application, 29(3), 232–240.

Itten Color Theory (1961)

Subjective color experience — Johannes Itten's The Art of Color codified the emotional and symbolic properties of twelve hues on the color wheel. Itten described warm-cool contrast as the most fundamental color opposition, with red-orange as the warmest and blue-green as the coolest. He linked hue position to psychological temperature, weight, and spatial forward/backward movement. While not as rigorous as psychophysical studies, Itten's framework remains the foundation for design education.

Reference: Itten, J. (1961). The Art of Color. Reinhold Publishing Corporation, New York.

Jzazbz Perceptual Color Space (Safdar, 2017)

HDR-optimised uniform appearance model — Jzazbz is a perceptual color space designed for high dynamic range (HDR) and wide color gamut (WCG) imaging. It uses absolute luminance (cd/m²) as input and applies a PQ (Perceptual Quantizer) non-linearity based on the Barten contrast sensitivity model. Jz correlates with perceived lightness with minimal hue-related lightness errors. Az/Bz axes represent chroma and hue with better uniformity than CIELAB. Used in this tool as the primary perceptual engine for warmth and emotion computations.

Reference: Safdar, M., Cui, G., Kim, Y.J. & Luo, M.R. (2017). Perceptually uniform color space for image signals including high dynamic range and wide gamut. Optics Express, 25(13), 15131–15151.

CAM16-UCS Colour Appearance Model (Li, 2017)

Successor to CIECAM02 — CAM16 models human color appearance under various viewing conditions (adapting luminance, background luminance, surround). It computes J (lightness), C (chroma), h (hue angle), M (colorfulness), s (saturation), and Q (brightness). The UCS (Uniform Color Space) variant maps J', a', b' uniformly for ΔE computation. CAM16 is preferred over CIELAB for cross-illuminant and cross-device accuracy. Used as the secondary model in this tool.

Reference: Li, C., Li, Z., Wang, Z., Xu, Y., Luo, M.R., Cui, G., Melgosa, M., Brill, M.H. & Pointer, M. (2017). Comprehensive colour appearance model (CAM16). Color Research & Application, 42(6), 703–718.

WCAG 2.1 & APCA Contrast

WCAG 2.1 (W3C, 2018) — Web Content Accessibility Guidelines define contrast ratio as (L1 + 0.05) / (L2 + 0.05) where L1 > L2 are WCAG relative luminances. AA requires ≥4.5:1 for normal text, ≥3:1 for large text. AAA requires ≥7:1. Despite widespread adoption, WCAG 2.x is known to mispredict perceptual contrast for dark colors due to incorrect gamma handling.

APCA — Advanced Perceptual Contrast Algorithm — Designed for WCAG 3.0, APCA uses a spatial frequency-based model and separate lightness curves for text and background. Lc (lightness contrast) values: ≥ 75 Lc for body text, ≥ 45 Lc for large text, ≥ 30 Lc for spot text. Better models human contrast sensitivity than WCAG 2.x, especially for light-on-dark combinations.

Formulas & Mathematics
Jzazbz Conversion (sRGB → Jzazbz)
Step 1 — sRGB → Linear RGB (sRGB EOTF):
C_lin = C_sRGB / 12.92   if C_sRGB ≤ 0.04045
C_lin = ((C_sRGB + 0.055) / 1.055)^2.4   otherwise

Step 2 — Linear RGB → Absolute XYZ (D65, 203 cd/m²):
Scale: X_abs = X * 203, Y_abs = Y * 203, Z_abs = Z * 203

Step 3 — XYZ → LMS (Dolby CAT):
| 0.41479 0.57960 -0.04963 |
|-0.20151 1.12249 0.05316 |
|-0.01166 0.26435 0.96797 |

Step 4 — PQ non-linearity (Barten model):
L' = ((c1 + c2*(L/10000)^n) / (1 + c3*(L/10000)^n))^m
c1=0.8359375, c2=18.8515625, c3=18.6875, n=0.159302, m=78.844

Step 5 — LMS' → Izazbz → Jzazbz:
Iz = 0.5*L' + 0.5*M' - 0.5*S'...
Jz = (1 + d)*Iz / (1 + d*Iz) - d0   (d=0.56, d0=1.629e-11)
Warmth Index Formula
H = HSL hue (0–360°), S = HSL saturation (0–1), L = HSL lightness (0–1)

warmHue = cos((H - 30) * π/180) -- peaks at 30° (orange)

Base warmth = 0.5 + 0.5 * warmHue

Warmth Index = Base + 0.2 * S * (1 - |2L - 1|) - 0.1 * (1 - S)

Clamp: warmth ∈ [0, 1]

Interpretation:
≥ 0.65 → Warm
0.35 – 0.65 → Neutral
< 0.35 → Cool
Emotion Score (Arousal-Valence)
Based on Valdez-Mehrabian (1994) regression approximation:

Pleasure ≈ 0.69*L - 0.02*S + 0.01*H_norm + 0.42
Arousal ≈ 0.22*S - 0.19*L + 0.08*H_peak + 0.31
Dominance ≈ 0.12*S + 0.06*L - 0.10*(H near purple) + 0.50

where L, S ∈ [0,1], H_norm = normalized hue position,
H_peak = peak activation function at red (0°/360°).

Final emotion label assigned from quadrant in arousal-valence plane:
High A + High V = Excited / Delighted
High A + Low V = Tense / Alarmed
Low A + High V = Calm / Relaxed
Low A + Low V = Depressed / Bored
Attention / Distinctiveness Index
Saturation-driven attention (simplified Treisman pop-out model):

Attention = 0.60 * S + 0.25 * (1 - |2L - 1|) + 0.15 * C_norm

where S = HSL saturation, L = HSL lightness,
C_norm = normalised chroma from OKLCh ∈ [0,1].

Interpretation:
≥ 0.70 → High attention (distinct, pop-out likely)
0.40 – 0.70 → Moderate
< 0.40 → Low (blends into background)
Brand Fit Score
Each context has a target (hue_center, S_min, L_range, chroma_min):

hue_score = max(0, 1 - |H - hue_center| / 60)
sat_score = S >= S_min ? 1.0 : S / S_min
lum_score = L in L_range ? 1.0 : falloff(L, L_range)

Brand Fit = 0.40*hue_score + 0.35*sat_score + 0.25*lum_score

Contexts and target hue centers:
brand-primary → flexible, weight trust + energy
cta → warm hues 10–50° or high-sat any hue
background → low S, high or mid L
warning → red-orange 0–30°
success → green 100–150°
info → blue 200–240°
WCAG 2.1 Relative Luminance & Contrast Ratio
Linearise (sRGB EOTF):
C_lin = C_sRGB / 12.92   if C_sRGB ≤ 0.04045
C_lin = ((C_sRGB + 0.055) / 1.055)^2.4   otherwise

Relative luminance:
L = 0.2126R_lin + 0.7152G_lin + 0.0722B_lin

Contrast ratio (L1 ≥ L2):
CR = (L1 + 0.05) / (L2 + 0.05)

AA: CR ≥ 4.5:1 (normal), CR ≥ 3:1 (large text)
AAA: CR ≥ 7:1 (normal), CR ≥ 4.5:1 (large text)
References & Citations

Emotion & Affect Models

[1] Russell, J.A. (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39(6), 1161–1178.

[2] Valdez, A. & Mehrabian, A. (1994). Effects of color on emotions. Journal of Experimental Psychology: General, 123(4), 394–409.

[3] Elliot, A.J. & Maier, M.A. (2014). Color psychology: Effects of perceiving color on psychological functioning in humans. Annual Review of Psychology, 65, 95–120.

Color-Emotion Scaling

[4] Ou, L.C., Luo, M.R., Woodcock, A. & Wright, A. (2004). A study of colour emotion and colour preference. Part I: Colour emotions for single colours. Color Research & Application, 29(3), 232–240.

[5] Ou, L.C., Luo, M.R., Sun, P.L., Hu, N.C. & Chen, H.S. (2012). Colour emotion and colour harmony: Part 2: A combined model. Color Research & Application, 37(4), 257–270.

[6] Itten, J. (1961). The Art of Color. Reinhold Publishing Corporation, New York.

Perceptual Color Models

[7] Safdar, M., Cui, G., Kim, Y.J. & Luo, M.R. (2017). Perceptually uniform color space for image signals including high dynamic range and wide gamut. Optics Express, 25(13), 15131–15151. (Jzazbz)

[8] Li, C., Li, Z., Wang, Z., Xu, Y., Luo, M.R., Cui, G., Melgosa, M., Brill, M.H. & Pointer, M. (2017). Comprehensive colour appearance model (CAM16). Color Research & Application, 42(6), 703–718.

[9] CIE 15:2004 — Colorimetry (3rd Edition). CIE Central Bureau, Vienna.

Cultural & Cross-Cultural Research

[10] Hupka, R.B., Zaleski, Z., Otto, J., Reidl, L. & Tarabrina, N.V. (1997). The colors of anger, envy, fear and jealousy: A cross-cultural study. Journal of Cross-Cultural Psychology, 28(2), 156–171.

[11] Madden, T.J., Hewett, K. & Roth, M.S. (2000). Managing images in different cultures: A cross-national study of color meanings and preferences. Journal of International Marketing, 8(4), 90–107.

[12] Aslam, M.M. (2006). Are you selling the right colour? A cross-cultural review of colour as a marketing cue. Journal of Marketing Communications, 12(1), 15–30.

Contrast & Accessibility

[13] W3C (2018). Web Content Accessibility Guidelines (WCAG) 2.1. World Wide Web Consortium.

[14] Myndex Research (2022–2026). Advanced Perceptual Contrast Algorithm (APCA) — WCAG 3.0 candidate. GitHub: Myndex/SAPC-APCA.

[15] Barten, P.G.J. (1999). Contrast Sensitivity of the Human Eye and Its Effects on Image Quality. SPIE Press.

About this tool

This tool computes emotion (arousal-valence via Valdez-Mehrabian regression), warmth index (Ou et al. hue mapping), saturation-driven attention, brand-fit for 8 contexts, and high-level cultural associations — all on-device, zero network. Perceptual computations use Jzazbz (Safdar 2017) and CAM16-UCS (Li 2017). Contrast is computed via WCAG 2.1 and APCA. Not a substitute for professional brand research or user testing.

Research & Visualization
Arousal-Valence Plane (Russell, 1980)

The current color is plotted in the Russell circumplex model. Each quadrant corresponds to an emotional cluster: excited (high arousal, positive valence), tense (high arousal, negative valence), calm (low arousal, positive valence), and bored/depressed (low arousal, negative valence).

Warmth Spectrum Visualisation

Position of the current color on the warm-cool hue spectrum. Derived from hue angle and modulated by saturation and lightness per the Ou et al. (2004) warmth scaling model.

Hue-Emotion Heatmap

Hue-angle mapped to emotion labels across the full spectrum (0–360°). Red arrow marks the current color's hue. Based on aggregated consensus from Itten (1961), Ou et al. (2004), and Elliot & Maier (2014).

Extended Cultural Associations
Run the Lab tab first to populate cultural data.
Research note: All charts render from the Lab tab's computed perceptual values. Switch to Lab, pick a color, then return to Research to view updated visualisations. Charts support double-click to fullscreen. All computation is on-device — zero network calls.