This study suggests that the soothing power of words has more to do with tonal cues to speech than semantic content. The researchers were able to tease this apart by having young children undergo a stressor and then had them in one of four conditions:
- No contact with mother
- Instant messaging with mother
- Partial contact with mother (talking over the phone)
- Full contact with mother (talking face to face)
They then measured the amount of salival cortisol (associated with stress) and urinary oxytocin (associated with forming and maintaining positive relationships) and compared the groups. In the end, they found that the no contact and instant messaging groups had significantly high cortisol levels, and significantly low levels of oxytocin (both relative to the partial and full contact groups).
If one thinks of humans as evolved creatures, these findings make sense. Stressors can come in a variety of different forms (e.g, failure to attain resources, failure to find/retain a mate, salient disease/physical safety threat, etc.) and having a way to relieve that stress would be adaptive. I wouldn’t say that speech came about for this purpose, but, given that speech is incredibly useful for other things (like allowing us to benefit from others’ knowledge) and so became standard issue for humans, it is plausible that further complexity could build off of that foundation. The finding that semantic content does not matter much is surprising, but makes sense given that young children also have stressors and it takes them some time to learn a language.
I would like to see if similar results would be found in adults, considering that some stressors can be explained away with semantic content.