Can feelings of loneliness and depression be adaptive for adolescents? Yes, suggests the Evolutionary Theory of Loneliness: initially, loneliness activates depressive responses for self-preservation. In turn, these feelings prepare adolescents for reaffiliation, thereby reducing loneliness (H1a: short-term balancing feedback loop). If this fails, loneliness and depressive symptoms may become long-term traits that exacerbate each other over time (H1b: long-term reinforcing feedback loop). Therefore, the short-term balancing feedback loop between loneliness and depressive symptoms may buffer against their long-term increases (H2: across-timescale influence). We tested these hypotheses in Dutch adolescents (Mage_Wave1 = 12.9, SDage_Wave1 = 0.7, 53% female) using six-wave, half-yearly panel data (N = 774; 2017–2021) and 7-day, 1.5-hourly experience sampling data (nsubsample = 84; mid-2021). Residual dynamic structural equation modeling revealed reinforcing feedback loops at both short-term (1.5-hourly) and long-term (half-yearly), supporting H1b but not H1a. Bayesian latent change score modeling supported H2: Adolescents who felt more depressed an hour after heightened loneliness showed smaller half-yearly increases in trait loneliness. However, this buffering effect was not predicted by the hourly depressed-to-loneliness relation, nor did either hourly relations predict half-yearly changes in depressive symptoms. Our findings suggest that feeling depressed shortly after loneliness may protect adolescents from long-term loneliness.

Economists have long known that timescale matters in that the structure ofdecisions as to the relevant time horizon, degree of time aggregation,strength of relationship, and even the relevant variables differ by timescale. Unfortunately, until recently it was difficult to decompose economictime series into orthogonal timescale components except for the short orlong run in which the former is dominated by noise. Wavelets are usedto produce an orthogonal decomposition of some economic variables by timescale over six different timescales. The relationship of interest is thatbetween money and income, i.e., velocity. We confirm that timescaledecomposition is very important for analyzing economic relationships. Theanalysis indicates the importance of recognizing variations in phase betweenvariables when investigating the relationships between them and throwsconsiderable light on the conflicting results that have been obtained in theliterature using Granger causality tests.