consumption, a second non-dimensional parameter is of
importance. This is the fraction of the total surface area being in
contact with the oral surfaces defined as AC/S. Physical melting is
proportional to both parameters as can be obtained through
dimensional analysis. For simplicity we consider the product of
both parameters as the main influencing factor of physical melting,
which results to:
The lower the ratio AC/V, the slower is the melting. Therefore,for each shape, we measured volume and amount of surface area being initially in contact with the oral surfaces. The ratio AC/V was then computed and compared to the sensory attribute “melting”. The result is shown in Fig. 7 for a subset of the shapes. The correlation
between the parameter representing physical melting and perceived melting is very good (r ¼ 0.94). The shapes Catstongue 1 and Catstongue 2 were removed from this assessment, because panellists’ comments indicated a fracture of the shapes at a very early state of consumption, meaning that AC/V changes fundamentally
with breakage.Further, a better prediction of melting, in the specific conditions of the test, would have to take into consideration the dynamic process of eating. While melting chocolate progressively loses its initial shape under the action of different parts of the mouth and of saliva, the contact area between the chocolate and the mouth evolves during consumption. Moreover, part of the food material is swallowed during consumption, meaning that the volume of
product to consider when calculating the heat conduction also evolves over time. This further model refinement needs to be investigated in future research.Differences in melting perception were expected to lead to
differences in flavour perception. Indeed, the aroma compounds entrapped in the food matrix are released in the mouth during the melting phase and conveyed to the olfactory system through the retronasal pathway. If melting speed is different from shape to shape, the flavour release and consequently the overall perception
of the flavour should differ. Accordingly, quick melting should induce a faster aroma release and thus a stronger cocoa flavour. This is contradictory to our findings showing that cocoa intensity was negatively correlated with the perceived melting. However, it may be that a slower melting rate contributes to a stronger cocoa flavour
because stimuli are released in the mouth over a longer period of time. This is what would suggest a recent study from Ruijschopet al. (2011) that showed that an increased duration of oral processing led to higher cumulative retro-nasal aroma stimulation per gram consumed. In order to investigate further the link between cocoa flavour
and perceived melting and especially test whether slower melting was associated to specific dynamic patterns of cocoa perception, timeeintensity dynamic sensory measurements were employed. Based on the outcomes of the sensory profiling, only four extreme shapes were selected: the Wing, the Sail, the Round and the Rectangle.