**4. Discussion**

This study sought to examine the effects of WPI and SPI on cognitive function and vitamin B12 status in low B12 individuals. Cognitive function was measured for a variety of important domains using a robust cognitive assessment battery. Overall, there were no effects of WPI on cognitive function, which generally remained stable across visits during this treatment. On the contrary, SPI appeared to have benefits for the cognitive domains of reasoning speed and reaction time in females only. Given the age of participants in this sample (45–75 years) and the presence of this effect true for females only, this finding possibly reflects the impact of soy isoflavones on cognitive function in post-menopausal women [40]. A recent meta-analysis concluded that soy isoflavones exert a positive effect on a broad range of cognitive abilities in females and that this effect is robust in women as young as 60 years [41].

Despite the absence of a main treatment effect for WPI in terms of its ability to improve cognitive performance, positive findings were evident regarding changes in biometric measures of vitamin B12 status during WPI treatment and changes in cognitive function. Specifically, increases in active B12 and serum folate were associated with corresponding improvements in some cognitive domains. Reductions in homocysteine levels were also consistently related to improvements in cognitive performance, including working memory. Whilst there were no clear treatment effects linking WPI specifically to improved cognitive function, the changes observed in these key biochemical markers and the relation of these to improved cognition were consistent with the broader research literature regarding the role of B12, folate and homocysteine in helping to maintain good cognitive function [12].

The finding of direct associations between changes in biochemical and cognitive measures despite the absence of any effect directly attributable to WPI likely reflects the choice of the control treatment. More specifically, although SPI is a useful comparison given its different nutritional profile to WPI in terms of levels of B-vitamins and folate, the isoflavones present can have an impact on cognitive function, most notably in females [40]. Therefore, both treatments were arguably 'active' in relation to their ability to affect cognitive performance and the absence of any interaction effect favouring WPI does not preclude potential benefits of this for improving cognition as a result of its demonstrated influence on vitamin B-status. Thus, future studies will need to consider their choice of control treatment more carefully if the primary outcome concerns cognitive function.

In addition to using a relatively active control treatment, this study employed a relatively short intervention duration (eight weeks). Prior research has proposed that interventions involving otherwise healthy participants, as used herein, require durations as long as 2-to-5 years in order to demonstrate an impact on cognitive function [42]. Thus, it has been suggested that such studies might benefit from recruiting individuals at-risk of enhanced rates of cognitive decline. Given the relationships observed herein between changes in B12 related markers and improved cognitive performance, future studies should consider the potential for dairy-based supplements to slow the rate of decline in individuals with quantifiable Age-Associated Memory Impairment (AAMI). These individuals are particularly interesting given the absence of underlying pathology (e.g., dementia) explaining their decline and the fact that they do not ye<sup>t</sup> meet criteria for MCI [43].

The potential utility of using WPI or SPI for extended periods of time as a means of supplementing diets must be balanced against any potential negative impacts of these. For example, dairy (WPI) has been linked with self-reported gu<sup>t</sup> discomfort and subsequent avoidance of such products [44]. Question marks also remain in relation to soy and its potential impact on endocrinology [45], as well as the presence of contaminants such as aluminium in soy products [46].

In conclusion, although this intervention did not demonstrate a direct effect of WPI supplementation on cognitive function, the results provide some support to the notion of vitamin B12 and folate status having a role in supporting cognitive health. Specifically, important vitamin B12 and folate related markers were improved as a result of WPI supplementation and these changes were correlated with improvements in cognitive performance. However, the findings from this study are limited to similarly aged adults with subclinical vitamin B12 deficiency and further research is needed to explore how these findings may relate to other groups and populations.

**Author Contributions:** Conceptualisation, I.T.Z., V.S.D. and M.F.; Formal analysis, I.T.Z., D.H. and K.B.; Methodology, I.T.Z.; Project administration, I.T.Z., K.B. and V.S.D.; Supervision, I.T.Z.; Writing—original draft, I.T.Z., D.H. and K.B.; Writing—review & editing, D.H., V.S.D. and M.D. All authors read and approved the final manuscript.

**Funding:** This research was funded by the Dairy Research Institute, USA (National Dairy Council).

**Acknowledgments:** We thank Anne McGuffin (Clinical Trials Manager) and Lindy Lawson (Clinical Nurse) for their role in the planning and collection of samples and other data during the intervention trial.

**Conflicts of Interest:** The authors declare no conflict of interest.
