Loneliness has been considered a major challenge since long before the pandemic. Changes caused by the COVID-19 pandemic included modifications in social communications and activities. Thus, it was expected that loneliness would increase during the pandemic. The first studies of loneliness during the pandemic revealed inconsistent results. We hypothesized that physical isolation led to changes in the quality of relationships; thus, loneliness trends could be different from those predicted. For our study we used methods to measure loneliness: the Social and Emotional Loneliness Scale (SELSA-S) for Adults and Older Adults; the Multidimensional Inventory of Loneliness Experience; and demographic data. Participants were middle-aged and older middle-aged adults (n = 457) aged 35-59 (M = 45.5, SD = 6.88, 35.4% males). Participants came from two studies: Study 1 consisted of 280 participants aged 35-59 (M = 44.8; SD = 6.93; 29.6% males), the study was conducted before the pandemic in late 2019; participants in Study 2 were adults (n = 177) aged 35-59 (M = 46.5; SD = 6.68; 44.6% males), data were collected in the fall of 2020. The results did not confirm increase in loneliness; moreover, participants reported lower scores of loneliness in some domains. Regression analyses showed that general experience of loneliness was predicted by different loneliness characteristics in pre-pandemic and pandemic age groups. We found some similar mechanisms that were activated within different situations. Our results confirmed the complex nature of loneliness, they argue that pandemic effects were not limited to increase in loneliness and that the mechanism of loneliness can adjust to environmental factors.