| # | Study | Outcome | Metric | Value | 95% CI | k | N | I² | Notes |
|---|---|---|---|---|---|---|---|---|---|
| 1 | Dikeçligil & Gottfried, 2024 | Variance in perceived guaiacol intensity explained by OR10G4 genotype | % | 15 | Mainland et al. 2014, cited secondarily in the review; valence >10% | ||||
| 1 | Dikeçligil & Gottfried, 2024 | Variance in perceptual space explained by pleasantness | % | 30 | Khan et al. 2007 (Dravnieks Atlas, 159 odorants), cited secondarily | ||||
| 1 | Dikeçligil & Gottfried, 2024 | Prediction of 19 semantic descriptors from cheminformatic features (population) | r | 0.41 | DREAM Challenge, Keller et al. 2017; individual r = 0.21 | ||||
| 4 | Torske et al., 2022 | ALE peak in left amygdala, all odours vs odourless | ALE | 0.127 | 108 | 1744 | Cluster k = 1,163 voxels; −22, 0, −20; ALE metric not in the protocol’s list | ||
| 4 | Torske et al., 2022 | Aversive > Pleasant, right insula/OFC | t | 3.67 | k = 525 voxels; 36, 20, −8; P95 threshold | ||||
| 4 | Torske et al., 2022 | Food > Aversive, left parahippocampal/amygdala/entorhinal | t | 2.75 | k = 222; −14, −2, −26 | ||||
| 4 | Torske et al., 2022 | Pleasant > Aversive, left hippocampus/parahippocampal/OFC | t | 2.15 | k = 158; −20, −8, −22 | ||||
| 5 | Zhou et al., 2021 | Piriform–hippocampus vs auditory–hippocampus theta synchrony (iEEG) | t | 4 | 8 | t(7), P = 0.0052; 8/8 participants | |||
| 5 | Zhou et al., 2021 | Effect of region on left hippocampal connectivity (olfactory vs visual/auditory/tactile) | F | 25.8 | 25 | F(3,72), P = 1.92e-11; right F = 16.37 | |||
| 5 | Zhou et al., 2021 | Connectivity of 7 targets (olfactory vs non-olfactory) with the hippocampus | F | 18.5 | 25 | F(6,144), P = 7.77e-16 | |||
| 5 | Zhou et al., 2021 | Anterior > posterior preference, temporal piriform | F | 15.67 | 25 | F(1,24), P = 0.00059 | |||
| 8 | Kurian et al., 2021 | Proportion of odorants antagonising at least one receptor | % | 50 | Pfister et al. 2020, ~800 odorants, cited secondarily | ||||
| 8 | Kurian et al., 2021 | Proportion of mouse ORs expressed (RNA-seq) | % | 98.9 | Ibarra-Soria/Saraiva, cited secondarily | ||||
| 11 | Kehl et al., 2024 | Mean amygdala firing across 15 odours vs standard valence scale (Toet 2020) | r | 0.56 | 15 | Spearman, P = 0.03; N = 15 odours | |||
| 11 | Kehl et al., 2024 | Hippocampal decoding vs correct identification rate across sessions | r | 0.5 | 27 | P = 0.0076; PC r = 0.14, amygdala r = 0.06 not significant | |||
| 11 | Kehl et al., 2024 | Amygdala: liked vs disliked odours (firing) | z | 2.9 | 130 | Wilcoxon, P = 0.004; N = 130 neurons | |||
| 11 | Kehl et al., 2024 | Proportion of PC neurons modulated by odour | % | 39.5 | 17 | ± 4.7% per session; N = 17 sessions | |||
| 11 | Kehl et al., 2024 | Repetition suppression in the amygdala over 8 presentations | z | -4.5 | 130 | P = 6.9 × 10⁻⁶ | |||
| 12 | Sagar et al., 2023 | RSA neural–perceptual similarity in OFC | r | 0.12 | 3 | Pooled Spearman, P < 0.0001; molecular rm = 0.077 | |||
| 12 | Sagar et al., 2023 | RSA fine-grained perceptual code in OFC | r | 0.083 | 3 | Raw rc = 0.063; rf > rc P < 0.0001 | |||
| 12 | Sagar et al., 2023 | RSA neural–perceptual similarity in the amygdala | r | 0.059 | 3 | P < 0.0001; rm = 0.050 | |||
| 13 | Sagar et al., 2026 | RSA valence in OFC | β | 0.569 | 3 | Standardised β, P < 0.001; β metric not in the protocol’s list | |||
| 13 | Sagar et al., 2026 | Trial-wise salience decoding in OFC | r | 0.157 | 4400 | N = trials; valence r = 0.145 | |||
| 13 | Sagar et al., 2026 | Aversive > appetitive pleasantness decoding in the amygdala | z | 4.584 | P < 0.001; PirC z = 4.263 | ||||
| 13 | Sagar et al., 2026 | Cross-decoding appetitive↔aversive in OFC (negative sign = salience code) | r | -0.068 | 2197 | P = 0.002 | |||
| 19 | Soudry et al., 2011 | Depression in olfactory dysfunction after acute rhinitis (Faulcon 1999) | % | 60 | “nearly 60%” had some degree of depression in the first months; N not stated | ||||
| 20 | Saive et al., 2014 | Odour recognition at short intervals (Engen & Ross 1973) | % | 73 | % correct | ||||
| 22 | Yang et al., 2025 | Difference in endpoints/voxels across 8 amygdala nuclei | F | 9.3057 | 25 | One-way ANOVA, df = 7, p = 6.6e-10 | |||
| 22 | Yang et al., 2025 | Endpoints of olfactory vs non-olfactory nuclei | t | 7.31 | 25 | df = 99, p = 7.1e-11 | |||
| 22 | Yang et al., 2025 | Number of unique streamlines, olfactory vs non-olfactory | t | 3.7 | 25 | paired, df = 49 (left/right separately), p = 5.4e-4 (Figure 3 caption gives p = 2.0e-5) | |||
| 23 | Noto et al., 2021 | k=2 clustering of olfactory nuclei, right hemisphere | z | 4.27 | 25 | permutation, P = 2.0e-6 | |||
| 23 | Noto et al., 2021 | k=2 clustering of olfactory nuclei, left hemisphere | z | 4.79 | 25 | permutation, P = 1.7e-6 | |||
| 23 | Noto et al., 2021 | k=3 clustering matching anatomical nuclei | z | 6.05 | 25 | minimum z | |||
| 24 | Buchanan et al., 2003 | Odour–name matching false alarms across 3 groups | F | 5.6 | 40 | F(2,37), p = 0.007, η² = 0.23 | |||
| 24 | Buchanan et al., 2003 | Response bias C for odour–name matching across 3 groups | F | 3.5 | 40 | F(2,37), p = 0.041, η² = 0.16 | |||
| 24 | Buchanan et al., 2003 | Odour–name matching d′, pooled lobectomy vs controls | t | 1.7 | 40 | t(38), p<0.05 | |||
| 24 | Buchanan et al., 2003 | Group × old/new interaction on familiarity scores | F | 6.2 | 40 | F(1,38), p = 0.017, η² = 0.14 | |||
| 24 | Buchanan et al., 2003 | Residual right amygdala volume vs name-matching false alarms | r | 0.42 | p<0.05; right hippocampus r = 0.46 | ||||
| 27 | Li & Wilson, 2023 | Proportion of piriform voxels shifting tuning to CS+ on day 9 (You 2022) | % | 20 | >20%; absent on day 1 | ||||
| 29 | Tong et al., 2014 | Improvement in discrimination learning, block 1 day 2 vs day 1 | t | 7.56 | t(8), p < 0.001; mice | ||||
| 32 | Hackländer et al., 2019 | Emotional intensity of odour-cued vs other-cued AMs | d | 0.47 | 10 | 9 experiments/10 data points; t(9) = 2.33, p = .022 | |||
| 32 | Hackländer et al., 2019 | Feeling of being brought back | d | 0.47 | 7 | t(6) = 2.36, p = .028 | |||
| 32 | Hackländer et al., 2019 | Less rehearsed (Rare) | d | 0.52 | 4 | t(3) = 3.54, p = .019 | |||
| 32 | Hackländer et al., 2019 | Pleasant valence of the memory | d | 0.35 | 10 | 8 experiments/10 points; t(9) = 3.43, p = .004 | |||
| 32 | Hackländer et al., 2019 | Vividness | d | 0.1 | 10 | t(9) = 2.02, p = .037; possibly overestimated | |||
| 32 | Hackländer et al., 2019 | Cue effectiveness (proportion of trials yielding an AM) | d | -0.5 | 8 | estimated d ≈ .50 against odours (M .63 vs .76; Wilcoxon Z = 2.52, p = .012); negative sign… | |||
| 32 | Hackländer et al., 2019 | Decade × modality interaction on the proportion of memories by age | F | 16.2 | 5 | F(2,8), p = .002, ηp² = .80 | |||
| 34 | Dejou et al., 2026 | Earliest childhood odour memories from events repeated > 5 times (human survey) | % | 73.1 | 647 | 13.4% repeated 2–5 times; 13.4% once | |||
| 34 | Dejou et al., 2026 | Still periodically re-encounter childhood odours (human survey) | % | 82 | 647 | ||||
| 34 | Dejou et al., 2026 | Pleasantness of reported vs unreported odour descriptors | t | 2.01 | 146 | t(98.9), p = 0.024; 51 vs 95 descriptors | |||
| 35 | Yang et al., 2021 | Theta phase separation between sequence positions, correct vs incorrect trials, anterior hippocampus | t | 7.32 | 6 | t5, p < 0.001; PCx t5 = 5.97, p = 0.002 | |||
| 35 | Yang et al., 2021 | PCx–AH theta PLV during replay | t | 8.68 | 5 | t4, p < 0.001 | |||
| 35 | Yang et al., 2021 | Proportion of replay before correct vs incorrect responses | % | 75 | 7 | 75% vs 53%; Fisher p < 1e-4 | |||
| 35 | Yang et al., 2021 | Correlation of replay with % correct (bootstrap) | r | [0.12; 0.15] | only CI95 [0.12, 0.15] reported | ||||
| 36 | Szychowska et al., 2025 | Olfactory vs visual spatial memory | SMD | -0.48 | [-0.77; -0.19] | 7 | 71.09 | Glass, random effects; z = −3.22, p = 0.0013; PI95 [−1.15, 0.19] | |
| 37 | Huang et al., 2026 | OB volume vs olfactory function | r | [0.15; 0.93] | 23 | range of r across studies (not a CI); 20/23 significant | |||
| 37 | Huang et al., 2026 | Reduced OBV and/or OSD vs controls | % | 90 | 50 | 45/50 studies | |||
| 37 | Huang et al., 2026 | DTI indices vs olfactory function | r | [0.31; 0.52] | 3 | range of r for the 3 significant studies (not a CI); 1 not significant | |||
| 37 | Huang et al., 2026 | Cortical thickness near the OS vs TDI in congenital anosmia (Ottaviano 2015) | r | -0.5 | 1 | p = 0.001 | |||
| 38 | Smitka et al., 2012 | Right hippocampal volume vs odour threshold | r | 0.21 | 117 | Pearson, p = 0.02 | |||
| 38 | Smitka et al., 2012 | Right hippocampal volume vs odour threshold, controlling for age | r | 0.18 | 117 | partial r(114), p = 0.049 | |||
| 41 | Keller et al., 2007 | Androstadienone EC50, S84N vs RT (in vitro) | % | EC50 nearly 20 times lower than RT (ratio ~20×); absolute values only in the figure | |||||
| 41 | Keller et al., 2007 | WM allele frequency (R88W/T133M) | % | 15.4 | 412 | allele frequency 0.154 | |||
| 43 | Stephan et al., 2012 | Single-cell response termination τ (IBMX) | MD | 466 | ms; 612 ± 53 (KO) − 146 ± 13 (WT); calculated from the note’s data | ||||
| 43 | Stephan et al., 2012 | Ratio τ(Na⁺-free)/τ(Ringer) | % | WT 4.5 ± 0.6-fold vs KO 1.08 ± 0.03-fold, P < 0.001 | |||||
| 43 | Stephan et al., 2012 | % of cells firing to the second pulse at a 0.25 s interval | % | 45 | WT ~45%; KO 0% (up to a 1 s interval) | ||||
| 44 | Reisert & Zhao, 2011 | % of WT OSNs firing to the second pulse at a 0.25 s interval | % | 50 | ~50%; Song 2008; n = 19–49 OSNs, χ² P = 0.007 | ||||
| 46 | Gottfried, 2010 | % of piriform neurons (APC/PPC) responding sub-additively to mixtures | % | [40; 60] | about 40–60% sub-additive; 20–40% supra-additive (rodent imaging);… | ||||
| 47 | Wilson & Sullivan, 2011 | Proportion of LEC units responding to odours | % | 35.2 | 128 | 45/128 (~one third); % calculated from the note’s data | |||
| 49 | Howard et al., 2009 | Prediction of inter-individual perceptual structure from PPC patterns | r | 0.44 | 4 | F₁,₂₄ = 18.84; P = 0.001; leave-one-subject-out | |||
| 49 | Howard et al., 2009 | PPC Procrustes goodness-of-fit (neural vs perceptual) | % | 0.53 | 4 | goodness-of-fit (lower = better), outside the 95% CI of the shuffled distribution; actual metric is… | |||
| 49 | Howard et al., 2009 | Within- > between-odour correlation in OFC (Exp. 1) | t | 5.4 | 6 | T₅; P = 0.0029 | |||
| 49 | Howard et al., 2009 | Category classification in PPC (Exp. 2) | t | 2.34 | 4 | T₃; P = 0.050 (right at the threshold) | |||
| 50 | Zelano et al., 2011 | PPC template strength vs behavioural accuracy | r | 0.64 | 12 | P = 0.02; between individuals | |||
| 50 | Zelano et al., 2011 | Same−different target difference in PPC vs accuracy | r | 0.66 | 12 | P = 0.01 | |||
| 50 | Zelano et al., 2011 | APC encodes the target before the odour (same > different target) | t | 6.3 | 12 | T11; P < 0.0001 | |||
| 51 | Jiang et al., 2017 | Four-class decoding accuracy in PPC/amygdala | % | 46 | 4 | SD 7.3%; chance 25%; in 4 participants the PPC was the best electrode | |||
| 51 | Jiang et al., 2017 | % of trials with a theta peak in the inhalation phase (odour vs air) | % | 78.4 | 7 | air 55.6% ± 6.4%; Z = 2.366, p = 0.018 | |||
| 51 | Jiang et al., 2017 | Regression of theta on electrode position | F | 30.69 | F(1,240); r² = 0.21; p < 0.001 | ||||
| 53 | Grabenhorst et al., 2007 | Anterolateral OFC BOLD vs pleasantness ratings | r | 0.91 | 14 | [40, 52, −6], Z = 2.69, p = 0.042 | |||
| 53 | Grabenhorst et al., 2007 | Medial OFC BOLD vs pleasantness ratings | r | 0.88 | 14 | [4, 46, −14], Z = 2.41, p < 0.05 | |||
| 53 | Grabenhorst et al., 2007 | Piriform BOLD vs intensity | r | 0.86 | 14 | [−18, 4, −16]; with pleasantness r = −0.15, p = 0.60 | |||
| 53 | Grabenhorst et al., 2007 | Pleasantness ratings across the three odours | F | 41.8 | 14 | F(2,26), p ≪ 0.001 | |||
| 54 | Bao et al., 2019 | ERC grid-pattern strength vs prediction performance | r | 0.48 | 25 | robust regression, p = 0.03; one outlier > 3 SD | |||
| 54 | Bao et al., 2019 | Slope of the ERC pattern effect by angular deviation | t | -2.65 | 25 | t24, p = 0.014 | |||
| 55 | Bushdid et al., 2014 (with Meister 2015; Gerkin & Castro 2015) | Discrimination resolution (majority of participants) — critical % overlap | % | 51.17 | 26 | pairwise 57.43%; regression y = −0.81x + 91.45 | |||
| 55 | Bushdid et al., 2014 (with Meister 2015; Gerkin & Castro 2015) | Regression of discrimination rate on % overlap (per participant) | % | -0.81 | 26 | slope coefficient; intercept 91.45 | |||
| 56 | Lee et al., 2023 | % of molecules for which the GNN is closer to the panel mean than the median panellist | % | 53 | 323 | RF baseline 41% | |||
| 56 | Lee et al., 2023 | Proportion of chemical samples with significant odour-active impurities | % | 31.5 | [27.4; 35.6] | 95% CI; estimated from GC-MS/GC-O | |||
| 56 | Lee et al., 2023 | Accuracy on odd-one-out triplets (GNN vs RF) | % | 50 | 41 | RF 19%; binomial p < 0.01 | |||
| 56 | Lee et al., 2023 | Mean cross-validation AUROC of the GNN | AUC | 0.89 | 5000 | 5-fold CV, ~5,000 molecules | |||
| 58 | Sela & Sobel, 2010 | Proportion of people noticing manipulation of ambient odour (Lorig 1992) | % | 3.2 | 93 | 3/93; % calculated from the note’s data | |||
| 60 | Zald & Pardo, 1997 | Left amygdala rCBF vs unpleasantness | r | -0.51 | 20 | pooled 12 sulfide pairs + 8 UPSIT pairs; p < 0.05; lower score = more unpleasant | |||
| 60 | Zald & Pardo, 1997 | Left amygdala rCBF vs unpleasantness (first pair only) | r | -0.56 | 12 | p < 0.05 | |||
| 60 | Zald & Pardo, 1997 | Left OFC rCBF vs unpleasantness | r | -0.46 | 20 | p < 0.05 | |||
| 60 | Zald & Pardo, 1997 | Covariation of left OFC rCBF ↔ left amygdala | r | 0.65 | 20 | p < 0.005 | |||
| 60 | Zald & Pardo, 1997 | Left amygdala activation, sulfide vs odourless | z | 5.2 | 10 | Talairach (−34, −4, −11) | |||
| 60 | Zald & Pardo, 1997 | Right amygdala activation, sulfide vs odourless | z | 5 | 10 | Talairach (26, −1, −14) | |||
| 62 | Winston et al., 2005 | Amygdala BOLD: concentration × odour type interaction | F | 4.7 | 17 | df (2.0; 31.7); p = 0.017 | |||
| 62 | Winston et al., 2005 | Amygdala BOLD high − low, unpleasant odour | t | 4.1 | 17 | 0.23 % ± 0.058; p < 0.001 | |||
| 62 | Winston et al., 2005 | Amygdala BOLD high − low, pleasant odour | t | 2.3 | 17 | 0.14 % ± 0.061; p = 0.017 | |||
| 62 | Winston et al., 2005 | Amygdala BOLD high − low, neutral odour | t | 0.005 | 17 | 0.0001 % ± 0.057; p = 0.50 | |||
| 63 | Li et al., 2008 | Within-pair vs between-pair pattern correlation (method check, before conditioning) | f | 0.66 | p < 0.005; metric “f” as in the note, not in the standard list; N not established (… | ||||
| 64 | Krusemark et al., 2013 | Change in valence (neutral − negative) vs increase in anxiety | r | -0.61 | 14 | p < 0.05 | |||
| 64 | Krusemark et al., 2013 | Change in valence of neutral odours relative to negative odours | t | -3.03 | 14 | t(13); p < 0.01 | |||
| 64 | Krusemark et al., 2013 | Right pgACC BOLD vs increase in anxiety | r | 0.77 | 12 | p = 0.002; conjunction | |||
| 64 | Krusemark et al., 2013 | Total DCM modulation by neutral odours vs increase in anxiety | r | 0.6 | 12 | p = 0.04 | |||
| 64 | Krusemark et al., 2013 | Neutral-odour detection RT (advantage over air) after vs before anxiety | t | 2.1 | 14 | t(13); p = 0.05; −457 → −314 ms | |||
| 65 | Iravani et al., 2021 | RSA OB gamma – valence (250–325 ms) | r | 0.6 | 19 | controlling for intensity; p < 0.010 | |||
| 65 | Iravani et al., 2021 | RSA OB beta – valence (≈800 ms) | r | 0.65 | 19 | p < 0.014 | |||
| 65 | Iravani et al., 2021 | PAM at 0.5 s, unpleasant vs pleasant odour difference | t | 3.24 | [0.06; 0.23] | 44 | t(174); p < 0.001; CI is for the PAM difference; BF10 = 3.32; preregistered | ||
| 65 | Iravani et al., 2021 | Leaning away from negative odours | t | 2.47 | 44 | t(174); p < 0.007 one-sided | |||
| 66 | Root et al., 2014 | PI for TMT avoidance when OB axons are inactivated in CoA (light on) | PI | -7.9 | 11 | vs −65 ± 3.4 without light; PI = (P − 25)/0.25, non-standard metric | |||
| 66 | Root et al., 2014 | PI for 2-phenylethanol attraction when inactivated in CoA | PI | -8.3 | 6 | vs +48 ± 3.2 without light | |||
| 66 | Root et al., 2014 | PI on optogenetic activation of TMT-tagged CoA neurons | PI | -49 | 6 | ± 4.6 | |||
| 66 | Root et al., 2014 | PI on optogenetic activation of 2-phenylethanol-tagged CoA neurons | PI | 42 | 4 | ± 6.4 | |||
| 67 | Chrea et al., 2009 | Canonical correlation of emotion factors ↔ odour quality | r | 0.83 | p < 0.001; NC1 | ||||
| 67 | Chrea et al., 2009 | Correlation of Relaxation ↔ Refreshment factors (GEOS) | r | 0.83 | 2256 | CFA latent correlation, Table 6 | |||
| 71 | de Groot & Smeets, 2017 | Fear chemosignalling, all outcomes (after removing outliers, r_high) | g | 0.36 | [0.31; 0.41] | 1240 | 2.8 | Q(18) = 18.52, p = 0.422; abstract figure; N = 1,240 is the N of the full analysis set | |
| 71 | de Groot & Smeets, 2017 | Fear chemosignalling, overall, r_low | g | 0.36 | [0.27; 0.46] | 1240 | 31.1 | Q(20) = 29.03, p = 0.087 | |
| 71 | de Groot & Smeets, 2017 | Fear chemosignalling, overall, r_high | g | 0.34 | [0.26; 0.42] | 1240 | 55.6 | Q(20) = 45.09, p = 0.001 | |
| 71 | de Groot & Smeets, 2017 | Fear chemosignalling, after removing outliers, r_low | g | 0.4 | [0.32; 0.48] | 0 | Q(18) = 16.48, p = 0.559 | ||
| 73 | Mujica-Parodi et al., 2009 | Left amygdala BOLD STRESS > EXERCISE (original) | t | 4.8 | 16 | Z = 3.68; p(SVC) = 0.02; MNI (−16, −10, −18) | |||
| 73 | Mujica-Parodi et al., 2009 | Left amygdala BOLD STRESS > EXERCISE (replication) | t | 5.21 | 16 | Z = 3.88; p(SVC) = 0.008; MNI (−27, −6, −12) | |||
| 73 | Mujica-Parodi et al., 2009 | Slope σ of the psychometric function for angry-face discrimination (stress vs exercise) | F | 8.3 | 14 | p = 0.01; increase of ~43 % | |||
| 73 | Mujica-Parodi et al., 2009 | Donor cortisol before vs after skydiving | t | 7.15 | 40 | 0.229 → 0.584 µg/dL; p < 0.001 | |||
| 74 | de Groot et al., 2020 | Axillary sweat volume by fear-intensity group (linear contrast) | F | 24.85 | 36 | F(1,35); p < 0.001; η² = 0.42 | |||
| 74 | de Groot et al., 2020 | Composite fear score (CFS) vs sweat volume | r | 0.6 | 36 | Spearman; p < 0.001 | |||
| 74 | de Groot et al., 2020 | Leave-one-out classification of fear intensity | % | 83.33 | 36 | 30/36, vs chance 33 % | |||
| 74 | de Groot et al., 2020 | Decay of the PID signal over time | eta2_partial | 0.53 | RM-ANCOVA F(34,68) = 2.23, p = 0.003; non-standard metric | ||||
| 76 | Zelano et al., 2016 | RT for fearful-face recognition, inhalation vs exhalation (nasal breathing) | t | -2.64 | 21 | t(20); p = 0.016; ~28 ms | |||
| 76 | Zelano et al., 2016 | Route × emotion × phase interaction on RT | F | 4.77 | 38 | F(1,36); p = 0.036 | |||
| 76 | Zelano et al., 2016 | Retrieval accuracy, inhalation vs exhalation (nasal breathing) | t | 2.85 | 11 | t(10); p = 0.017 | |||
| 76 | Zelano et al., 2016 | Correlation of piriform slow oscillations with respiration, nasal breathing (P7) | r | 0.64 | p = 0.002; mouth breathing r = 0.09 (n.s.) | ||||
| 76 | Zelano et al., 2016 | Amygdala delta power during inhalation vs fearful-face RT (P8) | r | -0.4 | p = 0.05; 24 trials | ||||
| 78 | Bagur et al., 2021 | % freezing, OBX vs sham (significant blocks) | NR | 0.77 | note only states “effect size 0.77 and 0.97”, metric type not stated; RM-ANOVA group p = 0.038 | ||||
| 78 | Bagur et al., 2021 | % freezing, OBX vs sham (second block) | NR | 0.97 | metric type not stated in the note | ||||
| 78 | Bagur et al., 2021 | % freezing, ChR2 stimulation of OB at 13 Hz vs no stimulation | NR | 2.08 | p = 0.016; metric type not stated in the note | ||||
| 78 | Bagur et al., 2021 | 4 Hz power vs micro-movements | r | -0.56 | p = 2.3e−54 | ||||
| 78 | Bagur et al., 2021 | Proportion of dmPFC neurons phase-locked to OB 4 Hz during freezing | % | 45 | 100 | vs ~23 % during activity; χ² = 9.82, p = 0.0017; N = neurons | |||
| 79 | Daniels & Vermetten, 2016 | Prevalence of odour-triggered panic attacks in the past month (refugees, Hinton 2004) | % | 45 | 100 | cited study; 58 % of these had flashbacks | |||
| 79 | Daniels & Vermetten, 2016 | Proportion with ≥50 % reduction in PTSD severity after HOC (Abramowitz & Lichtenberg 2010) | % | 58 | 36 | cited study; no control group | |||
| 79 | Daniels & Vermetten, 2016 | Olfactory bulb volume after childhood maltreatment vs controls (Croy 2013) | % | 80 | cited study; “~80 %” | ||||
| 80 | Rasch et al., 2007 | location memory (spatial), odour-cue SWS vs vehicle | g | 0.69 | [0.19; 1.18] | 18 | Exp.1, 7.5 h sleep; recalculated by Hu 2020; Z=2.717, p=0.007 | ||
| 80 | Rasch et al., 2007 | finger tapping, SWS | g | 0.08 | [-0.52; 0.37] | 18 | p=0.734; note records it as positive but the CI is skewed negative — minus sign possibly lost | ||
| 80 | Rasch et al., 2007 | location memory, REM | g | 0.05 | [-0.51; 0.4] | 17 | Exp.3, p=0.813; sign possibly lost | ||
| 80 | Rasch et al., 2007 | finger tapping, REM | g | 0.21 | [-0.67; 0.25] | 17 | p=0.366; g lies outside the CI → almost certainly −0.21 (minus sign lost) | ||
| 80 | Rasch et al., 2007 | TMR SWS vs wake (fMRI exp) | g | 0.64 | [0.05; 1.22] | 12 | Exp.5, 1.5 h sleep in the scanner, p=0.033; flagged as an outlier by Hu | ||
| 82 | Shanahan et al., 2018 | vmPFC reactivation ↔ memory benefit (over time) | r | 0.7 | [0.42; 0.86] | 18 | 90% CI; FIR correlation peak | ||
| 82 | Shanahan et al., 2018 | vmPFC, cued categories only | r | 0.61 | 18 | r(16)=0.61, p=0.01 | |||
| 82 | Shanahan et al., 2018 | posterior fusiform, cued categories only | r | 0.72 | 18 | r(16)=0.72, p<0.001 | |||
| 82 | Shanahan et al., 2018 | vmPFC cluster, reactivation–memory | t | 7.53 | 18 | t(16), p_FWE=0.01, MNI [6,46,−12] | |||
| 82 | Shanahan et al., 2018 | less forgetting for the cued category | z | 1.7 | 18 | Wilcoxon, p=0.04 one-tailed | |||
| 83 | Hu et al., 2020 | TMR overall, all measures | g | 0.29 | [0.21; 0.38] | 212 | 2004 | 71 | 91 experiments; Z=6.711 |
| 83 | Hu et al., 2020 | TMR after trim-and-fill | g | 0.18 | [0.06; 0.3] | 17 studies added, p=0.003 | |||
| 83 | Hu et al., 2020 | TMR three-parameter selection model | g | 0.13 | [0.06; 0.21] | p=0.001 | |||
| 83 | Hu et al., 2020 | TMR with odours | g | 0.32 | [0.15; 0.5] | 32 | 372 | equivalent to sounds; Q(2)=0.688 | |
| 83 | Hu et al., 2020 | TMR during SWS | g | 0.27 | [0.2; 0.35] | 174 | 1471 | 70 experiments | |
| 83 | Hu et al., 2020 | TMR during N2 | g | 0.32 | [0.04; 0.6] | 13 | 165 | 6 experiments, p=0.026 | |
| 83 | Hu et al., 2020 | TMR during REM | g | 0.06 | [-0.31; 0.18] | 15 | 142 | p=0.616; CI skewed negative → true g possibly −0.06 | |
| 83 | Hu et al., 2020 | TMR during wakefulness | g | 0.07 | [-0.09; 0.23] | 30 | 366 | p=0.394 | |
| 83 | Hu et al., 2020 | TMR for skill learning (NREM) | g | 0.51 | [0.37; 0.65] | 23 | 229 | 10 experiments | |
| 83 | Hu et al., 2020 | TMR for language learning (NREM) | g | 0.4 | [0.14; 0.65] | 13 | 158 | 9 experiments | |
| 83 | Hu et al., 2020 | TMR for emotional memory (NREM) | g | 0.1 | 12 | 97 | n.s., p=0.366 | ||
| 83 | Hu et al., 2020 | TMR for fear memory (NREM) | g | 0.02 | 4 | 97 | n.s.; disregarding direction g=0.44, p=0.004 | ||
| 84 | Chu & Downes, 2002 | number of new detailed sentences in phase 2, 4 conditions | χ² | 16.72 | 40 | df=3, p<0.001; congruent odour 120 vs label 71 | |||
| 84 | Chu & Downes, 2002 | pain ratings (absolute change), 3 groups | χ² | 19.32 | 42 | Kruskal–Wallis df=2, p<0.001 | |||
| 84 | Chu & Downes, 2002 | pleasantness ratings, 3 groups | χ² | 17.52 | 42 | Kruskal–Wallis df=2, p<0.001 | |||
| 87 | Yeshurun et al., 2009 | memory for first vs second association on day 8 | F | 6.4 | 16 | F(1,15), p<0.023; 50.5% vs 34.5% | |||
| 87 | Yeshurun et al., 2009 | unpleasant first association vs second | t | 3.53 | 16 | t(15), p<0.003; 56.3% vs 28.8% | |||
| 87 | Yeshurun et al., 2009 | left hippocampus, memory for first vs second association | F | 9.69 | 16 | F(1,14), p<0.007, odours only | |||
| 87 | Yeshurun et al., 2009 | day-1 activity predicts memory after 1 week (odours) | F | 14.6 | 16 | F(1,14), p<0.002 | |||
| 88 | Gottfried et al., 2004 | right posterior piriform, hits > CR | z | 3.34 | 15 | p<0.001 uncorrected; MNI 30,3,−12 (sign inferred from anatomy) | |||
| 88 | Gottfried et al., 2004 | right anterior hippocampus, hits > CR | z | 2.83 | 15 | p<0.005 uncorrected; y,z signs inferred from anatomy | |||
| 88 | Gottfried et al., 2004 | pleasant vs neutral hits | F | 6.7 | 15 | F(1,14), p<0.05; 0.89 vs 0.84 | |||
| 89 | Dahmani et al., 2018 | odour identification ↔ 4/8 VM errors, spatial learners | r | -0.57 | [-0.77; -0.33] | 24 | bootstrap BCa 95% CI, partial correlation (sex) | ||
| 89 | Dahmani et al., 2018 | odour identification ↔ 4/8 VM errors, whole group | r | -0.37 | [-0.52; -0.21] | 57 | |||
| 89 | Dahmani et al., 2018 | odour identification ↔ % landmarks found (wayfinding) | r | 0.26 | [0.06; 0.45] | 57 | |||
| 89 | Dahmani et al., 2018 | right hippocampal volume ↔ odour identification (spatial) | r | 0.32 | [0.03; 0.6] | 23 | |||
| 89 | Dahmani et al., 2018 | left mOFC thickness ↔ odour identification (spatial) | r | 0.33 | [0.02; 0.64] | 23 | |||
| 89 | Dahmani et al., 2018 | odour scores, mOFC+ vs controls | MD | -3.7 | [-6.02; -1.12] | 29 | mOFC+ n=9 (1 case excluded from the odour analysis) vs 20 controls; bootstrap difference | ||
| 90 | Igarashi et al., 2014 | proportion of dCA1 cells phase-locked to LEC 20–40 Hz, T1→T5 | % | 17.4 | from 7.9% to 17.4%, P<0.005 | ||||
| 90 | Igarashi et al., 2014 | proportion of LEC cells phase-locked to dCA1, T1→T5 | % | 24.4 | from 13.3% to 24.4%, P<0.005 | ||||
| 90 | Igarashi et al., 2014 | LEC spike phase-locking with learning | F | 21 | F(4,383), P<0.001 | ||||
| 91 | Lee et al., 2021 | proportion of successful learning sessions, fan-cell inhibition vs controls | % | 25 | 10 | controls 90.9%; binomial p=4.1×10⁻¹⁰ | |||
| 91 | Lee et al., 2021 | proportion of correct sessions, dopamine inhibition vs controls | % | 54.3 | 8 | controls 92.4%; binomial p=2.5×10⁻¹¹ | |||
| 91 | Lee et al., 2021 | fan cells expressing D1R | % | 42 | 2 | 185/441 cells, n=2 mice | |||
| 92 | Sullivan et al., 2000 | direction of odour–shock learning by age (group × age) | F | 9.29 | F(2,54), P<0.001 | ||||
| 92 | Sullivan et al., 2000 | maintenance of preference PN9→PN10 | F | 17.94 | F(1,33), P<0.001 | ||||
| 92 | Sullivan et al., 2000 | amygdala 2-DG, training × age | F | 3.37 | F(2,35), P<0.05 | ||||
| 93 | Moriceau & Sullivan, 2006 | odour–shock learning PN12–15, with vs without the mother | F | 25.563 | 32 | F(3,28), P<0.0001; N inferred from df | |||
| 93 | Moriceau & Sullivan, 2006 | corticosterone into the amygdala with the mother present → avoidance | F | 35.362 | F(2,20), P<0.0001 | ||||
| 93 | Moriceau & Sullivan, 2006 | 2-DG in the lateral amygdala nucleus | F | 27.94 | F(4,20), P<0.0001 | ||||
| 94 | Schaal et al., 2000 | negative facial expressions to anise odour, exposed vs unexposed | d | -1.72 | [-2.906; -0.527] | 1 | 24 | recalculated by PLoS ONE 2023; p≤0.005 | |
| 94 | Schaal et al., 2000 | head orientation towards anise odour | d | 1.29 | [0.324; 2.25] | 1 | 24 | PLoS ONE 2023; p=0.009 | |
| 94 | Schaal et al., 2000 | mouthing (Schaal 2000 + Wagner 2019 pooled) | d | 1.035 | [0.164; 1.906] | 2 | p=0.02; not specific to this paper | ||
| 94 | Schaal et al., 2000 | head orientation, pooled 4 studies of flavour exposure via the maternal diet | d | 1.24 | [0.56; 1.91] | 4 | 40 | PLoS ONE 2023; homogeneous; p<0.001 | |
| 94 | Schaal et al., 2000 | negative expressions, pooled 4 studies | d | -0.87 | [-2.39; 0.66] | 4 | 78.7 | PLoS ONE 2023; n.s. | |
| 96 | Devanand et al., 2015 | conversion to AD, UPSIT Q1 vs Q4 | HR | 10.33 | [4.55; 23.49] | 757 | discrete-time survival model | ||
| 96 | Devanand et al., 2015 | conversion to AD per UPSIT point (lower) | HR | 1.099 | [1.067; 1.131] | 757 | unadjusted; after covariates 1.072 (1.036–1.109) | ||
| 96 | Devanand et al., 2015 | conversion to AD per UPSIT point, adjusted | HR | 1.072 | [1.036; 1.109] | 757 | sex, age, language, education, SRT-TR, function | ||
| 96 | Devanand et al., 2015 | cognitive decline per UPSIT point | RR | 1.067 | [1.04; 1.095] | 732 | after covariates 1.065 (1.034–1.095) | ||
| 96 | Devanand et al., 2015 | cognitive decline per UPSIT point, people without MCI | RR | 1.096 | [1.06; 1.135] | 498 | after covariates 1.089 (1.050–1.130) | ||
| 96 | Devanand et al., 2015 | AUC of UPSIT predicting cognitive decline | AUC | 0.638 | 732 | SRT-TR 0.513; p<0.001 | |||
| 96 | Devanand et al., 2015 | AUC of UPSIT predicting AD | AUC | 0.65 | 757 | SRT-TR 0.73; combined 0.77 | |||
| 97 | Olofsson et al., 2021 | 5-year mortality risk, anosmia vs normal | HR | 3 | Pinto 2014 via review: “3 times higher”; approximate metric | ||||
| 97 | Olofsson et al., 2021 | mortality risk per additional correctly identified odour | % | -6 | Wilson 2011 via review, ~4 years | ||||
| 100 | Ball et al., 2020 | CMAI agitation (Ballard 2002, lemon balm) | MD | -11.1 | [-20; -2.2] | 1 | 71 | cluster RCT, 8 care homes, moderate certainty | |
| 100 | Ball et al., 2020 | CCMAI agitation (Lin 2007, lavender) | MD | -5.13 | [-13.21; 2.95] | 1 | 70 | not significant | |
| 100 | Ball et al., 2020 | NPI (Ballard 2002) | MD | -15.8 | [-24.4; -7.2] | 1 | 71 | moderate certainty | |
| 100 | Ball et al., 2020 | NPI (Burns 2011, lemon balm) | MD | 2.8 | [-5.84; 11.44] | 1 | 63 | moderate certainty, no benefit | |
| 100 | Ball et al., 2020 | NPI (Fujii 2008, lavender) | MD | -9 | [-17.89; -0.11] | 1 | 28 | low certainty | |
| 100 | Ball et al., 2020 | NPI (Lin 2007, lavender) | MD | -6.64 | [-10.85; -2.43] | 1 | 70 | low certainty | |
| 100 | Ball et al., 2020 | NPI (Takahashi 2020, cedar) | MD | -4.26 | [-9.46; 0.94] | 1 | 36 | CI includes 0 | |
| 100 | Ball et al., 2020 | QoL Blau (Burns 2011) | MD | 19 | [-23.12; 61.12] | 1 | 63 | not significant | |
| 100 | Ball et al., 2020 | MMSE (Fujii 2008) | MD | 1 | [-4.19; 6.19] | 1 | 28 | not significant | |
| 100 | Ball et al., 2020 | ADAS-cog (Takahashi 2020) | MD | -0.36 | [-6.6; 5.88] | 1 | 36 | not significant | |
| 100 | Ball et al., 2020 | Barthel ADL (Burns 2011) | MD | -0.5 | [-1.79; 0.79] | 1 | 63 | not significant | |
| 100 | Ball et al., 2020 | Caregiver burden J-ZBI (Takahashi 2020) | MD | -6.27 | [-12.29; -0.25] | 1 | 36 | low certainty | |
| 101 | Donelli et al., 2019 | Anxiety, all scales (inhaled lavender vs all controls) | g | -0.73 | [-1; -0.46] | 24 | 1682 | 85 | MA6; excluding 3 outliers g −0.54 [−0.73; −0.36] I² 65% |
| 101 | Donelli et al., 2019 | Anxiety, all scales (inhaled, only RCTs not at high RoB) | g | -0.64 | [-0.82; -0.47] | 535 | 0 | most robust estimate | |
| 101 | Donelli et al., 2019 | Anxiety, high-anxiety situations (inhaled) | g | -0.83 | 78 | excluding outlier g −0.67 I² 55%; mild anxiety not significant | |||
| 101 | Donelli et al., 2019 | HAM-A (Silexan 80 mg vs placebo) | MD | -2.9 | [-4.86; -0.95] | 5 | 1173 | 74 | MA1; excluding Kasper 2010 I² 28% |
| 101 | Donelli et al., 2019 | Zung SAS (Silexan 80 mg vs placebo) | MD | -2.62 | [-4.84; -0.39] | 3 | 451 | 17 | MA2 |
| 101 | Donelli et al., 2019 | STAI-S (inhaled vs usual care/none) | MD | -5.99 | [-9.39; -2.59] | 12 | 901 | 95 | MA3; excluding high RoB −3.67 p=0.05 I² 53% |
| 101 | Donelli et al., 2019 | STAI-T (inhaled vs usual care/none) | MD | -8.14 | [-14.44; -1.84] | 4 | 196 | 91 | MA4; excluding Mirbastegan −4.81 [−8.32; −1.31] I² 59% |
| 101 | Donelli et al., 2019 | Systolic blood pressure (inhaled vs none/sham) | MD | -1.91 | [-3.32; -0.51] | 6 | 398 | 0 | mmHg; r=0.5; 7 comparisons; inconclusive |
| 101 | Donelli et al., 2019 | Anxiety, all scales (lavender massage) | g | -0.66 | [-0.97; -0.35] | 6 | 448 | 61 | MA7; cannot be separated from the massage |
| 102 | Kasper et al., 2014 | HAMA (Silexan 160 mg vs placebo) | MD | 4 | [1.9; 6.1] | 1 | 256 | positive sign = greater reduction with Silexan; FAS 121+135; p<0.001 one-sided | |
| 102 | Kasper et al., 2014 | HAMA (Silexan 160 mg vs placebo) | d | 0.5 | 1 | 256 | Cohen’s d | ||
| 102 | Kasper et al., 2014 | HAMA (Silexan 80 mg vs placebo) | MD | 2.8 | [0.9; 4.8] | 1 | 270 | FAS 135+135; p=0.003 | |
| 102 | Kasper et al., 2014 | HAMA (Silexan 80 mg vs placebo) | d | 0.37 | 1 | 270 | Cohen’s d | ||
| 102 | Kasper et al., 2014 | HAMA (paroxetine 20 mg vs placebo) | d | 0.21 | 1 | 267 | not significant in the FAS | ||
| 102 | Kasper et al., 2014 | HAMA (Silexan 160 mg vs paroxetine) | MD | 2.8 | [0.7; 4.9] | 1 | 253 | FAS | |
| 102 | Kasper et al., 2014 | HAMA response rate ↓≥50% (160 mg vs placebo) | % | 60.3 | 1 | 121 | placebo 37.8% | ||
| 104 | Lin et al., 2019 | Sleep quality (aromatherapy vs control) | g | 1.103 | [0.813; 1.393] | 31 | 91.216 | Egger p=0.004, Begg p=0.008 — publication bias; estimate may be inflated | |
| 105 | Kiecolt-Glaser et al., 2008 | PANAS positive affect, time × odour | F | 4.43 | 1 | 56 | F(6,240), p<0.001; lemon > lavender, water | ||
| 105 | Kiecolt-Glaser et al., 2008 | Norepinephrine, time × odour | F | 2.35 | 1 | 56 | F(10,456), p=0.01 | ||
| 105 | Kiecolt-Glaser et al., 2008 | Candida DTH by odour | F | 4.47 | 1 | 54 | F(2,92), p=0.014; water > lavender; 2 participants excluded for lacking DTH | ||
| 105 | Kiecolt-Glaser et al., 2008 | Negative affect decreasing over time | OR | 2.8 | 1 | 56 | time effect only, p=0.001 | ||
| 107 | de Araujo et al., 2005 | Pleasantness of test odour: cheddar label vs body odour label | t | 6.68 | 1 | 12 | t(11); difference of 0.76 points on the −2..+2 scale | ||
| 107 | de Araujo et al., 2005 | Pleasantness of clean air: cheddar label vs body odour label | t | 4.1 | 1 | 12 | t(11), p<0.001 | ||
| 107 | de Araujo et al., 2005 | rACC/mOFC BOLD correlated with pleasantness | z | 4.35 | 1 | 12 | MNI [16,46,−4], p<0.05 corrected | ||
| 109 | Perl et al., 2016 | NREM delta power after vs before odour | % | 9.79 | 1 | 27 | SD 18.0; t(26)=3.22, P<0.005 | ||
| 109 | Perl et al., 2016 | NREM slow-spindle power after vs before odour | % | 5.16 | 1 | 27 | SD 7.81; t(26)=3.91, P<0.001 | ||
| 109 | Perl et al., 2016 | K-complexes per trial after vs before odour (N2) | t | 0.14 | 1 | 34 | t(33), P>0.55; BF01 = 6.61 supporting H0; note gives 95% CI −0.104 to −0.004 (inconsistent… | ||
| 110 | Hummel et al., 2009 | Proportion with clinically meaningful TDI improvement (>6 points), training group | % | 28 | 1 | 40 | vs 6% in the non-training group (n=16); from a secondary source | ||
| 111 | Sorokowska et al., 2017 | TDI after olfactory training | g | 1.1 | [0.459; 1.734] | 11 | 879 | Q = 172.73; I² NR | |
| 111 | Sorokowska et al., 2017 | Identification | g | 0.83 | [0.264; 1.402] | 13 | 1005 | Q = 200.66 | |
| 111 | Sorokowska et al., 2017 | Discrimination | g | 0.89 | [0.498; 1.298] | 10 | 788 | Q = 57.06 | |
| 111 | Sorokowska et al., 2017 | Threshold | g | 0.34 | [0.103; 0.569] | 12 | 951 | Q = 32.37 | |
| 111 | Sorokowska et al., 2017 | TDI, post-infectious subgroup | g | 1.52 | vs 0.45 in other groups; Q = 4.34, p = 0.04 | ||||
| 111 | Sorokowska et al., 2017 | Identification, post-infectious + post-traumatic subgroups | g | 1.54 | vs 0.29 | ||||
| 112 | Pieniak et al., 2022 | Proportion with clinical improvement after 6 months of OT (Qiao 2019) | % | 41.7 | 1 | 1.7% after 1 month; 26.7% after 3 months | |||
| 112 | Pieniak et al., 2022 | Proportion with clinical improvement after 56 weeks of OT (Konstantinidis 2016) | % | 71 | 1 | vs 37% without intervention | |||
| 113 | Kattar et al., 2021 | Clinically meaningful TDI improvement (>5.5 points), OT vs control | OR | 2.77 | [1.67; 4.58] | 4 | post-viral olfactory dysfunction; I² NR | ||
| 114 | Kohli et al., 2016 | BDI–TDI correlation in patients with olfactory dysfunction | r | -0.349 | 3 | 260 | weighted by sample size, p<0.001 | ||
| 114 | Kohli et al., 2016 | SS Identification: depression vs controls | MD | -0.41 | 360 | calculated from 12.57 vs 12.98 (152/208); p<0.0001 | |||
| 114 | Kohli et al., 2016 | SS Threshold: depression vs controls | MD | -0.47 | 291 | calculated from 6.31 vs 6.78 (122/169); p=0.0005 | |||
| 114 | Kohli et al., 2016 | Prevalence of depression (BDI) in anosmia after upper respiratory infection | % | 76 | 1 | 25 | Jung 2014; CRS 40% | ||
| 116 | Spangenberg et al., 1996 | Store evaluation (scented vs unscented) | F | 5.51 | 1 | 298 | F(6,291), p<0.0001 | ||
| 116 | Spangenberg et al., 1996 | Approach behaviour/intentions (multivariate) | F | 3.13 | 1 | 298 | F(12,285), p<0.001 | ||
| 116 | Spangenberg et al., 1996 | Intention to visit the store | MD | 0.74 | 1 | 298 | 4.20 → 4.94; p=0.003 | ||
| 116 | Spangenberg et al., 1996 | Perceived shopping time | % | -13 | 1 | 298 | 660.8 s vs 576.0 s (≈ −85 s); p=0.021; % calculated by the note’s reader | ||
| 117 | Roschk & Hosseinpour, 2020 | Recall | r | 0.177 | 29 | 1299 | adjusted r; +11.3% | ||
| 117 | Roschk & Hosseinpour, 2020 | Spending | r | 0.051 | 52 | 8150 | ns; +3.2% | ||
| 117 | Roschk & Hosseinpour, 2020 | Shopping satisfaction | r | 0.144 | 40 | 3095 | +9.2% | ||
| 117 | Roschk & Hosseinpour, 2020 | Environmental quality | r | 0.122 | 105 | 6709 | +7.8% | ||
| 117 | Roschk & Hosseinpour, 2020 | Product evaluation | r | 0.121 | 132 | 6878 | +7.7% | ||
| 117 | Roschk & Hosseinpour, 2020 | Time elusiveness | r | 0.111 | 13 | 1336 | +7.1% | ||
| 117 | Roschk & Hosseinpour, 2020 | Purchase intention | r | 0.105 | 49 | 2849 | +6.7% | ||
| 117 | Roschk & Hosseinpour, 2020 | Mood – activation | r | 0.097 | 75 | 6469 | +6.2% | ||
| 117 | Roschk & Hosseinpour, 2020 | Mood – valence | r | 0.088 | 120 | 8099 | +5.6% | ||
| 117 | Roschk & Hosseinpour, 2020 | Lingering | r | 0.078 | 33 | 2762 | +5.0% | ||
| 118 | Krishna et al., 2010 | Recall of product attributes by odour (Exp. 1) | F | 39.65 | 1 | 151 | F(2,141), η² = 0.36 | ||
| 118 | Krishna et al., 2010 | Recall of scented vs unscented tissues after 2 weeks (Exp. 2) | MD | 0.63 | 1 | 143 | 4.65 → 5.28; F(1,135) = 5.111, η² = 0.05 | ||
| 118 | Krishna et al., 2010 | Retention rate after 2 weeks, familiar odour | % | 76.8 | 1 | unscented 26.6%; novel odour 91.8% |
Scientific reference material, not medical advice. Figures are redrawn from published data, not copied from the original figures. GRADE and risk-of-bias ratings are the compilers’ own. Every number traces back to detailed reading notes on each paper.