Temporal Dominance of Sensations

1 Introduksi

1.1 Latar Belakang

Temporal Dominance of Sensations (TDS) is a dynamic sensory method measuring the evolution of perception over time of a consumed product — from first bite to aftertaste (Pineau et al., 2009). Unlike time-intensity (rate one attribute) or static descriptive (single snapshot), TDS asks panelists to select the currently dominant attribute from a list of 5–10 attributes, and change selection when dominance changes. Output: TDS curves per attribute — % panelists citing attribute $j$ as dominant at time $t$. Curves superimposed with chance level line ($1/n_{\text{attributes}}$) and significance line (binomial CI). Attributes above significance line at a given time = temporally dominant. Applications: flavor release profile design, cooking effect analysis, sensory benchmarking.

1.2 Tujuan

Receive TDS raw data: Panelist, Product, Replicate, Time_s, Attribute_dominant. Calculate TDS curves per attribute per product. Apply chance level $1/n_{\text{attr}}$ and significance line (binomial $\alpha = 0.05$). Display multi-product comparison with TDS difference plot. Audience: senior sensory researchers, R&D focusing on flavor release dynamics.

1.3 Posisi

TDS for dynamics multi-attribute perception. For single-attribute rating over time (time-intensity): separate module (coming). For static descriptive: QDA / Descriptive Profiling. For binary CATA: CATA + Cochran Q. For link descriptor → liking: Preference Mapping.

2 Metode

2.1 Dasar Teoretis

TDS dominance per attribute $j$ at time $t$:

$$\text{Dominance}_j(t) = \frac{N_j(t)}{N_{\text{total}}}$$

dengan $N_j(t)$ = jumlah panelis yang menyebut atribut $j$ dominan pada waktu $t$, $N_{\text{total}}$ = total panelis.

Chance level:

$$\text{Chance} = \frac{1}{n_{\text{attributes}}}$$

$$\boxed{\, P_{\text{sig}} = \text{Chance} + 1.645 \cdot \sqrt{\frac{\text{Chance} \cdot (1 - \text{Chance})}{N_{\text{total}}}} \,}$$

Atribut dengan dominance > $P_{\text{sig}}$ pada waktu tertentu = significantly dominant.

Time normalization (Pineau et al., 2009): normalize to $t / t_{\max,i}$ per panelist (0 to 1) before aggregation.

TDS difference curves for product comparison (Albert et al., 2012):

$$\Delta\text{Dom}_{A,B}(t) = \text{Dom}_A(t) - \text{Dom}_B(t)$$

Temporal indices (Pineau et al., 2012):

Dominance rate: max dominance per attribute.
Dominance duration: total time attribute dominant.
Onset time: first time attribute crosses significance.

2.2 Persamaan Inti

Dominance: $\text{Dom}_j(t) = N_j(t) / N_{\text{total}}$
Chance level: $\text{Chance} = 1/n_{\text{attributes}}$
Significance line: $P_{\text{sig}} = \text{Chance} + 1.645 \sqrt{\text{Chance}(1-\text{Chance})/N}$
TDS difference: $\Delta = \text{Dom}_A(t) - \text{Dom}_B(t)$

2.3 Asumsi & Batas Validitas

Asumsi	Konsekuensi jika dilanggar	Cara cek di SQalytics
Panel terlatih untuk TDS protocol	High variability	Pre-test session
Atribut lexicon agreed pre-test	Inconsistency	Lexicon validation
Replikasi $\geq 2$ per panelis	Curve noisy	Modul flag
$N \geq 30$ panelis	CI lebar	Konsumen panel $\geq 50$ rekomendasi
Time recording konsisten (interval $\leq 1$ s)	Curve resolution buruk	Catat sampling rate
Produk konsisten antar panelis (volume, T, vehicle)	Confounded	SOP serving

3 Cara Kerja

3.1 Step-by-Step di SQalytics

Buka Temporal Dominance of Sensations dari domain Sensori dan Riset Konsumen.
Muat raw data: Panelist, Product, Replicate, Time_s (atau normalized 0–1), Attribute_dominant.
Pilih time mode: raw seconds atau normalized.
Sesuaikan smoothing: window moving average (default 5% time).
Klik Run TDS Analysis.
Tinjau hasil: Tab TDS Curves, Tab Difference Curves, Tab Temporal Indices, Tab Heatmap.

Step 5 opsional — Publication Graph Studio (PGS). Temporal Dominance of Sensations sekarang menampilkan tombol 📰 Open in Publication Graph Studio. Anda dapat memilih Dominance Curves atau Peak Dominance; SQalytics akan mengirim figure temporal terpilih, ChartSpec, dan tabel dominansi bersih ke PGS.

3.2 Template Tabel Input + Contoh Data Sintetis

Kolom	Tipe	Wajib	Catatan
`Panelist`	category	✓	ID panelis
`Product`	category	✓	Nama produk
`Replicate`	numeric	✓	1, 2, 3
`Time_s`	numeric	✓	Detik (atau 0–1 normalized)
`Attribute_dominant`	category	✓	Nama atribut dominan saat itu

SYNTHETIC Asumsikan 50 panelis × 2 replicate × ~30 timepoints = 3000 baris.

Contoh data sintetis (8 baris + lanjutan):

Panelist	Product	Replicate	Time_s	Attribute_dominant
P01	Choco-A	1	2	Sweet
P01	Choco-A	1	5	Sweet
P01	Choco-A	1	10	Cocoa
P01	Choco-A	1	15	Cocoa
P01	Choco-A	1	20	Bitter
P01	Choco-A	1	25	Bitter
P01	Choco-A	1	30	Aftertaste
…

3.3 Contoh Luaran + Figure

TDS curve summary table (50 panelis, 5 atribut):

Time normalized	Sweet	Cocoa	Bitter	Creamy	Aftertaste
0.1	0.65	0.15	0.05	0.10	0.05
0.3	0.30	0.50	0.10	0.05	0.05
0.5	0.15	0.55	0.18	0.07	0.05
0.7	0.08	0.25	0.45	0.05	0.17
0.9	0.05	0.10	0.18	0.02	0.65

(Bold = above significance line)

Chance level = 1/5 = 0.20; Significance line = 0.20 + 1.645 × √(0.20·0.80/50) = 0.293.

Temporal indices table:

Attribute	Onset (norm time)	Peak time	Peak dominance	Duration above sig
Sweet	0.0	0.1	0.65	0–0.2 (20%)
Cocoa	0.2	0.4	0.55	0.2–0.6 (40%)
Bitter	0.6	0.7	0.45	0.6–0.8 (20%)
Aftertaste	0.8	0.9	0.65	0.8–1.0 (20%)
Creamy	never	—	0.10	0%

TDS curves Choco-A dengan temporal indices dan significance line — Gambar 1. **(a)** TDS curves Choco-A (n = 50 panelis, 5 atribut): Sweet dominan di awal (t = 0–0.2), Cocoa dominan di tengah (0.2–0.6), Bitter dan Aftertaste di akhir; garis putus-putus = chance (0.20), garis titik-titik = significance (0.293). **(b)** Temporal indices: dominance duration — Cocoa memiliki durasi terpanjang (40%), Creamy tidak pernah mencapai significance.

Kesimpulan ringkas: TDS Choco-A menunjukkan flavor evolution sequential: Sweet dominant pertama (0–20%), Cocoa transition tengah (20–60%), Bitter di akhir (60–80%), Aftertaste finishing (80–100%). Creamy never dominant — produk tidak smooth-textured atau panelis tidak detect. Pola TDS dapat dipakai untuk direct flavor design: bila kompetitor punya Cocoa dominant lebih lama, formulasi perlu modulate sweetness onset. Lanjut ke TDS difference curve vs kompetitor; atau ke multi-product TDS untuk benchmarking. Pakai Preference Mapping untuk link TDS curve attributes ke consumer liking.

4 Kesimpulan

4.1 Relevansi Real-World

Cokelat sequential flavor design.
Mint chewing gum flavor release duration.
Aftertaste profiling (wine/coffee).
Texture transition in composite foods.
Salt reduction strategy perception kinetics.

4.2 Where to Go from Here

QDA / Descriptive Profiling — paralel static analysis.
Preference Mapping — link temporal pattern ke liking.
Pineau et al. (2009) untuk introduction TDS.
ASTM E2454 + ISO 13299 untuk standar related descriptive.

⚙ Troubleshooting Cepat

No attribute crosses significance. Panel size kecil atau atribut lexicon broad — refine.

Multiple attributes overlap di same time. Wajar untuk complex products; itulah TDS information.

Time recording inconsistent. Auto-software recording lebih baik dari manual click.

i Riwayat Revisi

Tanggal	Revisi	Penulis
2026-05-12	Draft v2 publikasi (KaTeX TDS dominance + chance + significance line + APA Pineau/Albert/Schlich)	Claude

4 Referensi

Pineau, N., Schlich, P., Cordelle, S., Mathonniere, C., Issanchou, S., Imbert, A., Rogeaux, M., Etievant, P., & Koster, E. (2009). Temporal dominance of sensations: Construction of the TDS curves and comparison with time-intensity. Food Quality and Preference, 20(6), 450–455. https://doi.org/10.1016/j.foodqual.2009.04.005
Pineau, N., Goupil de Bouille, A., Lepage, M., Lenfant, F., Schlich, P., Martin, N., & Rytz, A. (2012). Temporal dominance of sensations: What is a good attribute list? Food Quality and Preference, 26(2), 159–165. https://doi.org/10.1016/j.foodqual.2012.04.004
Albert, A., Salvador, A., Schlich, P., & Fiszman, S. (2012). Comparison between temporal dominance of sensations (TDS) and key-attribute sensory profiling for evaluating solid food with contrasting textural layers. Food Quality and Preference, 24(1), 111–118. https://doi.org/10.1016/j.foodqual.2011.10.003
Schlich, P. (2017). Temporal dominance of sensations (TDS). In Time-dependent measures of perception in sensory evaluation (pp. 159–181). John Wiley & Sons.
Hutchings, S. C., Foster, K. D., Bronlund, J. E., Lentle, R. G., Jones, J. R., & Morgenstern, M. P. (2014). Mastication of heterogeneous foods: Peanuts inside two different food matrices. Food Quality and Preference, 31, 71–82. https://doi.org/10.1016/j.foodqual.2013.08.004