The direct conversion receiver demo uses two complex modulated input signals: the first is centered around 2.45GHz and it is representative of the desired signal, the second is centered around 2.453GHz (3MHz away from the desired signal) and it represents the blocker.
The local oscillator has the same carrier frequency of the desired signal, that’s why it performs a direct conversion to DC. The blocker is converted to 3MHz. The output ports in the demo only select the signal centered around 0 Hz. To see the down-converted blocker signal you need to add an extra carrier frequency equal to 3MHz and to select the option “complex baseband” on the output ports. In this way, two signals are generated by the output port; in order to super-impose them after the output ports, first you need to make sure that the sample time is sufficiently small to capture the entire spectrum (you need to resample the signals), second you need to perform a frequency shift of the blocker envelope. This can be done with a complex multiplication, as it is done (for example) in the digital transmitter block to plot the signals with the correct frequency offset.
In the given example, the output port is not probing the down-converted blocker signal. This is equivalent to assuming an ideal filter removing the undesired signal after the RF receiver. However, as the blocker has higher power than the desired signal, and the receiver is affected by finite IP2, the blocker will cause a non-negligible DC-offset that that it is actually correctly simulated and it affects the down-converted desired signal.
I hope this helps, with best regards,