First off, do as the first two posters indicated, and take care of the temp gauge's dastardly temperature compensator board. Bypass it with jumper as described in the linked references. We've owned four 240's made with that board and I had to bypass all of them.

Next topic:
There are two temp sensors, each in its own location. One for the gauge and one for the car's computer. The gauge sensor is under intake runner 2. The ECT sensor (to support the computer) may be under runners 3 + 4 as described by davitus. ECT failure would likely affect the computer and probably your gas mileage but will not turn the gauge into a liar, as it does not run the gauge.

One of our four cars that I did the bypass on turned out to have a faulty GAUGE sensor. ECT sensor was fine. But gauge sensor failure is far less common than temp comp board failure. I assume the comp boards will fail; it's that bad.

Bypass that d****d compensator board.

Then if you still have issues, test the gauge sensor's lead for a short. '86 is smack within the range of years of 'biodegradable' disintegrating wire insulation and there's a chance that wire is grounding out, which would cause the high temp readings. The sensor is a simple device; higher temps cause it to provide lower resistance. Grounding the wire = no resistance and so mimicks high temperature.
--
Sven: '89 245 NA, 951 ECU, expanded air dam, forward belly pan reaches oem belly pan, airbox heater upgraded, E-fan, 205/65-15 at 50 psi, IPD sways, no a/c-p/s belt, E-Codes, amber front corner reflectors, aero front face, quad horns, tach, small clock.