Touch combines two dimensions: sensory-discriminative (distinguishing texture, moisture, temperature) and emotional-affective (pleasantness, arousal). Almost any tactile experience unites both aspects, but studies often separate them. This work sits at the intersection of several fields: the study of pleasant touch, age-related changes in sensitivity, and the role of self-touch in self-care. Nearby in the archive are papers on how a second touch feels different, how aging affects perception, and how thermal imaging detects stress. This article stands out because it focuses specifically on self-touch (rather than passive touch by another person) and applies new data analysis methods.
Participants from two age groups (20–28 years and 65–75 years) applied cream to the skin of the left cheek or left forearm with the index finger of the right hand, then rated 40 sensation descriptors (26 sensory and 14 emotional) on a scale from 'does not describe at all' to 'describes very much.' Data were analyzed in three ways: classical factor analysis, network analysis based on graphical lasso, and nonlinear t-SNE clustering followed by KMeans. For network analysis, polychoric correlations and bootstrap were used to test stability.
Factor analysis revealed four sensory factors (texture, moisture, friction, temperature) and four emotional ones (sensuality, pleasure, calmness, negative affect). Young participants felt texture, moisture, and coolness more strongly, while older participants felt warmth and greater sensuality. A combined analysis of sensory and emotional descriptors yielded four factors: positive affect, texture, moisture, negative affect. Network analysis showed that older participants had significantly fewer significant connections between descriptors (25 versus 41 in young participants), but the proportion of mixed sensory-emotional connections was higher (32% versus 5%). Clustering revealed four clusters in each group, but with different compositions: in young participants—arousal, positive affect, texture-negative affect, moisture; in older participants—arousal-moisture, positive affect, negative affect, texture.
The work confirms that sensory sensitivity declines with age, but the emotional evaluation of touch becomes more positive. Unlike neighboring studies that used only classical linear methods, this one employed network analysis and clustering, which allowed seeing nonlinear relationships between sensory and emotional aspects. This is the first application of such methods to the Touch Perception Task, opening the way to a more complete modeling of tactile experience.
Network analysis and clustering methods can be extended to other participant groups (men, different age cohorts), a larger number of body sites, and various textures. Machine learning could be used to predict emotional reactions from sensory characteristics. Studying causal links between sensory and emotional aspects with advanced statistical models is promising.
The results may be useful in developing cosmetics and skin care products, especially for older people, where the emphasis should be on emotional aspects. They are also important for understanding the role of self-touch in maintaining a sense of self and well-being in older age.
Conduct a similar study with men and a wider age range. Include more body sites and different stimulus textures. Apply deep learning methods to analyze multidimensional touch data.
The work addresses a fundamental problem in the neuroscience of touch: how sensory and emotional components integrate into a single perception. Age-related changes in the balance of these components may be linked to peripheral (reduced density of mechanoreceptors) and central (changes in emotion processing) mechanisms, which remains an open question. Moreover, the role of C-tactile afferents in the age-related enhancement of positive affect is unclear.
In older people, touching their own skin evokes more positive emotions than in young people, even though they discriminate texture and moisture less well. It's similar to how, with age, we begin to appreciate not the details but the overall impression.
In William Gibson's novel Neuromancer, tactile sensations are transmitted through simulation, where emotional coloring can matter more than physical accuracy. The study shows that with age, the emotional component of touch indeed comes to the fore, echoing the idea that perception is shaped not only by sensory data but also by internal state.
Not published on arXiv. Here are its page, the PDF and the formalization: you can check it yourself, not only read our account.