The role of passive muscle tensions in a three-dimensional dynamic model of the human jaw

The role of passive muscle tensions in human jaw function are largely unknown. It seems reasonable to assume that passive muscle-tension properties are optimized for the multiple physiological tasks the jaw performs in vivo. However, the inaccessibility of the jaw muscles is a major obstacle to measuring their passive tensions, and understanding their effects. Computer modelling offers an alternative method for doing this. Here, a three-dimensional, dynamic model was used to predict active and passive jaw-muscle tensions during simulated postural rest, jaw opening and chewing. The model included a rigid mandible, two temporomandibular joints, multiple dental bite points, and an artificial food bolus located between the right first molars. It was driven by 18 Hill-type actuators representing nine pairs of jaw muscles. All anatomical forms, positions and properties used in the model were based on previously published, average values. Two states were stimulated, one in which all optimal lengths for the length-tension curves in the closing muscles were defined as their fibre-component lengths when the incisor teeth were 2 mm apart (S2), and another in which the optimal lengths were set for a 12.0 mm interincisal separation (S12). At rest, the jaw attained 3.6 mm interincisal separation in SZ, and 14.8 mm in S12. Activation of the inferior lateral pterygoid (ILP) and digastric (DG) muscles in various combinations always induced passive jaw-closer tensions, and compressive condylar loads. Maximum midline gape (from maximum bilateral co-activation of DG and ILP) was 16.2 mm in S2, and 32.0 mm in S12. When both model states were driven with muscle patterns typical for human mastication, recognizable unilateral and vertical "chopping" chewing cycles were produced. Both states revealed condylar loading in the opening and closing phases of mastication. During unilateral chewing, compressive force on the working-side condyle exceeded that on the balancing side. In contrast, during the "chopping" cycle, loading on the balancing side was greater than that on the working side. In S2, chewing was limited in both vertical and lateral directions. These results suggest that the assumptions used in S12 more closely approximated human behaviour than those in S2. Despite its limitations, modelling appears to provide a useful conceptual framework for developing hypotheses regarding the role of muscle tensions during human jaw function. (C) 1999 Elsevier Science Ltd. All rights reserved.