Thermal density (specific heat) of air is going to be ~1000x less than that of solid objects.
The specific heat of gypsum (the primary constituent of drywall) is 1.09 kJ/kg.K
For dry air it's 1.0 kJ/kg.K
Air's density is 1.225 kg/m3.
A 6m x 9m x 2.3m (20' x 30' x 7.5') room has a volume of about 130m^3, so a 10% exchange would be 13m^3, or 16kg.
That's about 15 kJ of heat energy per degree C, or roughly 0.0004 liter (0.0001 gallon) of heating oil equivalent.
The drywall would be (in feet) 20x7.5x2 + 30x7.5x2 + 20x30 ft^2 (I'll assume the floor is some perfect insulator for now, and that the room has no doorways), and 1/2 inch thick, or 1.6 m^3. That's about 3600 kg of gypsum, which has a heat capacity of about 4000 kJ per degree C, or about 0.1 liter (0.027 gallons) of heating oil equivalent.
If I'm doing my maths right.
Sources:
Specific heat of gypsum: http://www.engineeringtoolbox.com/specific-heat-solids-d_154...
Specific heat of dry air: http://www.engineeringtoolbox.com/air-specific-heat-capacity...
Density of gypsum: https://en.wikipedia.org/wiki/Gypsum
In winter in particular, the rate of through-roof air leaks is also greater when temperatures are warmer due to the stack effect. The author is completely wrong on this front - it's far better to let your house cool down and then re-heat.
The only time this isn't true is if you have a two-mechanism heating system such as a heat pump with resistance heat backup, where a large temperature swing invokes the more expensive resistance heater. The author doesn't.
I've got mote sensors in every room in my house, including the hall. When my kids leave the front door open in winter for a few minutes, as kids tend to do, I can see the temperature in the hall dive on the graphs. But close the door again, and the temperature is right back up very close to where it started in just a few minutes - even when the heating is off.