footnote on title page + move citation in discussion

sorry i cannot knit no more T_T

paper/manuscript.Rmd

@@ -298,7 +298,7 @@ Although yielding promising results, the sample size of this exploratory study i
 Another aspect to note is the strong reliance on self-reported measures as outcome variables of lie ability (in particular, the measure of answer confidence, but also the auto-questionnaires). This might conflate meta-cognitive abilities as well as dishonest answers. Although we tried to include more objective measures, such as RT (although it too was tied to the participants' conscious decision to press a key) and heart rate, future studies should attempt at measuring objectively the answer (lie or truth) quality, for instance by means of external examiners. Note that this is not a limitation *per se*, as it answers a slightly different question - what are the correlates of *objective* lying skills - rather than of deception self-confidence. 
 <!-- [**look for lit linking to what are commonly considered cues for lying**] -->
 
-Additionally to the limitations pertaining to the measure of lying ability, some also concern the measure of the predictor constructs, namely ToM and interoception. While we tried to include a behavioral task as well as a subjective questionnaire for each, it has to be underlined that they are notoriously difficult concepts to measure. In particular, objective interoceptive accuracy was assessed using the Heartbeat Counting Task (HCT). While the HCT used to be considered as a gold standard and remains one of the most commonly used measures [@desmedt2022does], concerns regarding its validity have been highlighted in several studies [@desmedt2018heartbeat; @desmedt2022does; @brener2016towards; @legrand2022heart]. Given increasing research efforts invested in the development of novel interoception tasks [@plans2021measuring; @legrand2022heart; @ponzo2021measuring], future works should further examine the relationship between interoception and lying ability using measures with better psychometric properties. 
+Additionally to the limitations pertaining to the measure of lying ability, some also concern the measure of the predictor constructs, namely ToM and interoception. While we tried to include a behavioral task as well as a subjective questionnaire for each, it has to be underlined that they are notoriously difficult concepts to measure. In particular, objective interoceptive accuracy was assessed using the Heartbeat Counting Task (HCT). While the HCT used to be considered as a gold standard and remains one of the most commonly used measures [@desmedt2022does], concerns regarding its validity have been increasingly highlighted in several studies as more research efforts are invested into developing novel interoception tasks [@desmedt2018heartbeat; @desmedt2022does; @brener2016towards; @legrand2022heart; @plans2021measuring; @ponzo2021measuring]. Future works should further examine the relationship between interoception and lying ability using measures with better psychometric properties. 
 
 Moreover, although the cognitive and affective components of ToM and empathy share overlaps in the current literature, and there is no consensus regarding how the two concepts should be delineated, recent evidence nonetheless suggests ToM and empathy are necessarily distinct constructs with separable underlying mechanisms [@kanske2015dissecting]. As such, future studies are warranted to further investigate the associations between ToM and lying ability using validated instruments sensitive to measuring ToM [such as the Theory of Mind Inventory, @hutchins2021development]. Furthermore, our application of feature reduction as a noise-elimination measure could have over-simplified the data. A more complex pattern of relationships, with different contributions of various subdimensions of ToM and interoception, could emerge with sufficient statistical power and valid measures.