The 2019 coronavirus pandemic (COVID-19) has had, and still has, a relevant effect on several aspects of people’s everyday life. Researchers have agreed on the fact that, since the pandemic began negative effects on both physical and psychological well-being, following full or partial lockdown periods, have been reported [1]. Moreover, some populations, such as immigrants, people with low socioeconomic status, homeless, older adults or people with chronic illness [2, 3], were considered more vulnerable than others.

Older people have been the primary target of most of the social distancing policies and regulations, mainly because they have a higher risk of presenting complications from COVID-19. Thus, they have often experienced and/or are still experiencing permanent isolation for several months, having limited social activities and thus usually perceiving greater feelings of loneliness [4]. This feeling of loneliness and boredom [5, 6] and relative physical inactivity [7] may have dramatic effects on individual physical and psychological well-being, as well as an increased risk of psychiatric disorders such as anxiety and depression [6] and an overall decline in cognitive functioning [8].

Interestingly, some studies have shown that cognitive reserve (CR), defined as the accumulation of brain resources during the entire lifespan [9, 10], can moderate the negative association between social isolation and cognitive decline [8]. This might also apply to the possible decrease in perceived psychological well-being [11]. More specifically, CR, according to different authors, denotes a cumulative improvement in neural resources due to the interaction of genetic and environmental variables that, in turn, might mitigate the effects of the neural decline caused by physiological aging or age-related diseases [10, 12]. This happens because the CR would allow people to optimize their performances through more adaptable functional brain processes [13], such as differential recruitment of brain networks (i.e., different brain areas or networks and interconnections) or alternative cognitive strategies [9, 14, 15]. Moreover, a direct relationship between these processes and healthy aging has been reported: the higher the CR, the better overall mental health an individual will achieve [13]. Several key factors influence the level of CR, including education, social networks, occupational history, number and type of leisure activities [12, 16]. Creativity has also been proved to be positively associated with CR and it hence can be considered a proxy for this construct [17, 18].

Starting from this evidence, it has been hypothesized that offering online-intervention activities connected to CR might not only increase participants’ CR itself, but also increase their overall well-being and decrease the feeling of social isolation of the older adults in a safe and fun way. In the present study we implemented the intervention by using a blog, designed to support participants’ collaboration and interactions. According to some studies, blogs might facilitate self-expression and self-reflection by promoting metacognition, because they allow participants to share knowledge and perspectives, to receive feedback, and to derive social support from each other [19, 20].

The aim of the present study was therefore to implement these theoretical premises by offering an online, blog-based, training to increase CR of a group of healthy aging people during the first lockdown period of COVID-19. We hypothesized that the training would have a positive effect on the participants, by increasing both their perceived well-being as well as their cognitive reserve level.

Analyses were run using IBM-SPSS 28 (Chicago, IL, USA) statistical package.

We first compared results from our intervention group (pre-post intervention), and then we proceeded to compare the scores of the Intervention group at Time 2 with the scores from our control group (who filled out the surveys at that time).

3.2 Effects of the Online Intervention on the Cognitive Reserve

As a second step, we focused on the effects of the intervention on participants’ CR level.

We run a paired-samples t-test to compare the scores from the CoRe-T at Time 1 and Time 2 for our intervention group. Results showed a significant increase between Time 1 and Time 2 in the overall CR scores (M${}_{\text{T1}}$ = 93.42; SD${}_{\text{T1}}$ = 13.55; M${}_{\text{T2}}$ = 106.51; SD${}_{\text{T2}}$ =15.30; t${}_{39}$ = –13.49; p 0.001; d = 0.89).

To assess changes in specific components of the CR that could have been affected differently by the intervention, we run a second paired-samples t-test to compare the scores of individual CR components (leisure activities and creativity, which are the dynamic components of CR as measured by Core-T) at Time 1 and Time 2 for our Intervention group (see Table 2 for mean scores and SD). The analysis returned a significant difference between Time 1 and Time 2 for both Leisure activities (t${}_{39}$ = –10.05, p 0.001; d = 0.63) and Creativity (t${}_{40}$ = –10.57, p 0.001; d = 0.63).

Table 2.Mean scores and standard deviations for main cognitive reserve (CR) sub-components.

	Time 1 (pre-test)	Time 2 (post-test)
Leisure activities	51.67 (9.02)	56.95 (7.63)

We then run an independent-samples t-test to compare intervention and control groups at Time 2, using the CoRe-T scores as dependent variables. The intervention group reported and gets significantly higher levels of cognitive reserve when compared to the quasi-equivalent group of peers living in the same area (M${}_{\text{Int}}$ = 106.51; SD${}_{\text{Int}}$ = 15.30; M${}_{\text{Contr}}$ = 93.54; SD${}_{\text{Contr}}$ = 13.15; t${}_{80}$ = –4.14, p 0.001; d = 0.91).

Finally, we ran another independent-samples t-test to examine more in depth where specific differences within the CR components that could have been affected by the intervention occurred. Distribution scores are reported in Fig. 1.

Fig. 1.

Box Plots of Cognitive Reserve Test (CoRe-T) scores for the Control and Intervention groups at Time 2.

Both the leisure activities component (t${}_{80}$ = –3.11; p 0.001; d = 0.69) and the creativity component (t${}_{80}$ = –2.90; p = 0.002; d = 0.64) of the CR were significantly different between the intervention and the control group.

Our main hypothesis, which led us to expect an increase in both perceived well-being and CR levels in a sample of healthy older people, after a 5-weeks of online blogging, was confirmed.

The intervention group reported significantly higher levels of well-being and got a higher score on CoRe-T after participating in the CR intervention. This difference emerged both when comparing the intervention group’s scores at Time 1 and 2, and when comparing the Intervention’s group’s scores with a quasi-equivalent control group matched by age, gender and education level. Previous research [23] reported how blogging appears to be helpful in meeting psychological and social needs for older adults, by providing a medium for this specific population to engage in activities linked to the well-being, like to focus on self-expression [24], engage in social and cognitive activities [25], and perceive to be part of a community [26].

Yet, we cannot forget how the role of CR for healthy aging has been considered a crucial variable in the field of aging neuroscience. The CR has been linked to a better quality of life [27], less cognitive decline [28] and lower incidence of psychiatric symptoms [29], so that interventions that target proxies of CR are highly recommended [27]. The blog-based intervention presented here addressed this specific issue by proposing different challenges inspired by CR proxies (see Supplementary Table 1); it was also unique because it was designed to specifically target creativity and divergent thinking, which are components linked to CR [17, 18] but, to our knowledge, have not been directly addressed in a CR intervention before.

It is also essential to stress the need to replicate these findings with different research paradigms (e.g., randomized controlled trials) and possibly longitudinal designs, to prove the effectiveness of online, blog-based interventions that focus on increasing participants’ CR.

Despite having statistically significant findings, there are some limitations that needs to be addressed. These limitations are mainly linked to a lack of a concurrent active control group (part of the effects can be ascribed to the activity of blogging per se), and a possible learning effect for the creativity items in the CoRe-T.

Regarding the lack of the control group, we tried to compensate by recruiting a group of individuals at Time 2, which has been living in the same area as the intervention group, hence sharing the same COVID-related restrictions. We matched them for age, gender and also educational level (which many times is considered a CR indicator by itself), assuming that this would allow us to make a relevant comparison about well-being and CR at Time 2, with most of the basic confounders that could have affected changes on the intervention group accounted for. Moreover, the intervention was relatively short in time, so that could also support our assumption that time related changes in the control group were most likely very minor This is not an ideal solution, and future studies should replicate our findings using a concurrent control group.

As for the learning effect for the creativity items of the Core-T test, we tried to compensate by accounting for the leisure activity component, which is not affected by a learning component, and should hence help having a more balanced post-test score. Finally, our analyses considered different subfactors of the more general flexibility of thought, and if the changes were always significant, the specific changes in each subfactor were different: meaning that either the learning effect was not so widespread or not all the components were affected.

The significant effect of an online blog-based intervention on both participants’ CR and perceived well-being reported in this paper has two main implications. On the one hand, it shows how even a short and relatively easy web-based intervention might have a significant impact on perceived well-being and CR, and, in turn might have positive effects on cognition and mental health of older people. This finding is in line with and support other recent researches that reported evidence of effectiveness of different types of tele-neuro-rehabilitation interventions [30, 31], also for older subjects [30]. On the other hand, our results highlight the potentiality of online training to reach, in both preventive and rehabilitative perspective, a large audience of participants (such as those who cannot access rehabilitation centers or other centralized healthcare services), not only during the COVID period, but also due to physical or economical restraints. These types of interventions are indeed generally considered promising healthcare tool, as they guarantee continuity of care over time and in space (from hospitals to patients’ home) and allow to increase the frequency and intensity of rehabilitation programs [32].

BC and SC designed the research study. BC, GF and SP performed the research. BC analyzed the data. BC and GF write the main parts of the manuscript. All Authors contributed to write the manuscript and approved the final manuscript.

Protocols and procedures for this research were approved by Champlain College’s IRB (COA IRB000173). Protocols and procedures were conducted according to the Declaration of Helsinki.

The authors would like to thank the Richmond Senior Center for promoting the research.

This research received no external funding.

The authors declare no conflict of interest. BC is serving as the Guest Editor for this special issue named “COVID-19 Related Restriction Measures and Their Impact on Cognition”. We declare that BC had no involvement in the peer review of this article and has no access to information regarding its peer review. Full responsibility for the editorial process for this article was delegated to CM.