How do I classify or spot a pixel, based on a specific reflectance values of the six Landsat Bands using Erdas?

This sounds like a homework assignment.

You will likely need to classify the image first. Probably in your case, you will do an unsupervised classification.

Here's a youtube video for performing an unsupervised image classification in Erdas Imagine.

Additionally, here's are some papers which may help you.

1. "Spatially targeting Culex quinquefasciatus aquatic habitats on modified land cover for implementing an Integrated Vector Management (IVM) program in three villages within the Mwea Rice Scheme, Kenya" Jacob et. al. use Erdas Imagine 8.7 to perform a "maximum likelihood unsupervised classification and generated three land cover classifications, rice field, fallow and built environment."

2. MONITORING OF PADDY RICE FIELDS IN AUSTRALIA USING ASTER DATA AND TM DATA Kitamura et. al. "estimated the areas of paddy rice fields in the study site in southern New South Wales State, Australia each year using TERRA-ASTER and Landsat-4,5 TM data acquired in the flooding season (about November) and the growing season (about February) from 2002/03 to 2009/10".

3. Mapping paddy rice agriculture in southern China using multi-temporal MODIS images Xiao et. al. "developed a paddy rice mapping algorithm that uses time series of three vegetation indices (LSWI, EVI, and NDVI) derived from MODIS images to identify that initial period of flooding and transplanting in paddy rice fields, based on the sensitivity of LSWI to the increased surface moisture during the period of flooding and rice transplanting".

Hope these help you.