An easy way to get a feel for this is to consider two liquids that form an ideal solution when mixed, and whose vapors form an ideal gas solution in the gas phase. Your two substances do not quite do this, but that won't detract from the analysis for the ideal case.
Let xa and xb represent the mole fractions of the two species in the liquid phase, and let ya and yB represent the mole fractions of the two species in the gas phase. Let P represent the total pressure, which in your situation is a constant. Let pa (T) be the equilibrium vapor pressure of pure a at temperature T, and pb (T) be the equilibrium vapor pressure of pure b at temperature T. Then for the ideal situation above,
P ya = pa xa
P yb = pb xb
If you add these two equations together, you get:
P = pa xa + pb (1 - xa)
Solving for xa, you then get:
xa= (P - pb) / (pa - pb)
For a given temperature, this equation gives the mole fraction of a in the liquid phase. The mole fraction of a in the gas phase is then given by:
ya = pa xa / P
In many cases, the ratio of the equilibrium vapor pressures pa / pb is fairly insensitive to temperature (i.e., assuming nearly matching heats of vaporization of the pure substances). This is called the assumption of constant "relative volatility." If we make this approximation, and apply it to the first two equations, we obtain:
ya/(1 - ya) = xa/ (1 - xa)(pa / pb)
According to this equation, ya would increase monotonically with xa under the constant relative volatility approximation.